UCSC has released v1.1 of the UCSC Proteome Browser. This updated version includes the following major enhancements:

• direct access to a Proteome Browser gateway via the "Proteome Browser" menu link on the Genome Browser home page. The Proteome Browser gateway page prompts the user for a protein ID or gene symbol, then directly starts up the Proteome Browser, eliminating the multiple Genome Browser steps required by the previous release.
• extended protein coverage that includes all proteins in the Swiss-Prot/TrEMBL databases, rather than just the human, mouse and rat protein sets included in the previous version.

In addition to providing direct access to the Proteome Browser, the v1.1 release preserves the existing tight coupling between the Proteome Browser and the Genome Browser for the human, mouse and rat genomes. Users may still navigate between the Genome Browser Known Genes track and the Proteome Browser for quick, easy access to the details of a gene's genomic and proteomic sequence structures.

The UCSC team who worked on the Proteome Browser update includes Fan Hsu, Robert Kuhn, Donna Karolchik and Tom Pringle. Please send comments or questions to our mailing list at genome@soe.ucsc.edu.

The UCSC Bioinformatics Group announces two seminars and hands-on workshops on the UCSC Genome Browser, presented by OpenHelix, a bioinformatics training, software testing and consulting company.

These introductory sessions are geared towards anyone with a basic knowledge of genomic and biological concepts who is interested in learning how to use the UCSC Genome Browser. No programming experience is required. The seminars will cover the topics necessary to learn how to effectively use the browser tool set, including basic Genome Browser functionality, searching and BLAT use, Table Browser use, creating and using custom annotation tracks, and an introduction to the Gene Sorter. Lectures will be accompanied by hands-on computer exercises conducted directly on the Genome Browser web site.

The first three-hour course will be held on Tuesday 9 November in the Washington, D.C. area. Two sessions will be offered: 9am-12pm and 1pm-4pm.

The second seminar will be held in the Raleigh/Durham, N.C. area on Wednesday, 10 November. Two sessions will be offered: 9am-12pm and 1pm-4pm.

For registration information, visit the OpenHelix website or call 1-888-861-5051. Academic, student, and early registration discounts are available.

We are pleased to announce the release of an enhanced version of the UCSC Table Browser. This new release sports several improvements and additions over the previous Table Browser, including:

• a new streamlined user interface
• support for generating filters that include fields from multiple tables, including those from non-positional tables
• an enhanced schema-viewing utility that displays all tables associated with a track, as well as all tables linked to a selected table
• the ability to restrict queries to include only data from ENCODE regions
• the addition of a GALA output option
• an option to save large output results sets directly to a file rather than displaying them in the Table Browser

The preliminary User's Guide will be enhanced in upcoming weeks to include examples of many common Table Browser queries generated by our users.

The new version of the Table Browser was produced by Jim Kent, Donna Karolchik, Heather Trumbower, Hiram Clawson, and Robert Kuhn, and incorporates code from the previous version written by Angie Hinrichs. We'd like to thank Mary Mangan and others on the OpenHelix staff for their feedback on early versions of this software.

The older version of the Table Browser will remain available for a limited time at http://genome.ucsc.edu/cgi-bin/hgText. Please send feedback and questions to our mailing list at genome@soe.ucsc.edu.

The Genoscope v7 Tetraodon nigroviridis genome assembly is now available in the UCSC Genome Browser and Blat server. This assembly, UCSC version tetNig1 dated Feb. 2004, is the result of a collaboration between Genoscope and the Broad Institute of MIT and Harvard.

The v7 assembly was constructed using the whole genome shotgun (WGS) approach, resulting in a sequence coverage of about 7.9X. The assembly contains 45,609 contigs and 25,773 scaffolds generated by the Arachne program and covers more than 90% of the genome. Additional linking data were used to build ultracontigs and to organize the assembly into chromosomes. Genoscope estimates the size of the Tetraodon genome to be about 385 Mb.

Downloads of the tetNig1 data and annotations may be obtained from the UCSC Genome Browser FTP server or Downloads page. These data have been freely provided by Genoscope before publication with specific conditions for use. The initial set of annotation tracks were generated by Genoscope and the UCSC Bioinformatics Group based on data provided Genoscope. Tetraodon gene predictions generated by Genoscope using GAZE will be available in the Genome Browser within a few weeks.

Many thanks to Genoscope and the Broad Institute of MIT and Harvard for this genome assembly. The UCSC team who produced this browser are Rachel Harte, Robert Kuhn, Donna Karolchik, and the Genome Browser sysadmin team. See the Credits page for a detailed list of the organizations and individuals who contributed to this release.

The UCSC Genome Bioinformatics Group has released a Genome Browser and Blat server for the Drosophila pseudoobscura Freeze 1 draft assembly (Aug. 2003). This assembly, UCSC version dp2, was produced by the Human Genome Sequencing Center (HGSC) at Baylor College of Medicine.

Freeze 1 is a whole genome shotgun assembly produced using Baylor HGSC's assembly engine, Atlas. The assembly, which provides approximately 7x coverage of the euchromatic portion of the genome, contains 759 scaffolds. The scaffold N50 size is 1,018,646 bp. The total scaffold size for this assembly is 139.3 Mbp, with an average size of 184,465 bp. Due to an assembly error, four large scaffolds "jumped" chromosomes. These have been split into "A" and "B" parts in the downloadable assembly files. See the gateway page for more information.

Baylor HGSC has provided a putative chromosome assignment for the majority of larger scaffolds (> 90% of unique sequence), based on conservation between the Muller elements.

Downloads of the dp2 data and annotations may be obtained from the UCSC Genome Browser FTP server or Downloads page. The initial set of annotation tracks were generated by UCSC.

Many thanks to the Baylor HGSC for providing the genome assembly data. The UCSC team who produced this browser are Angie Hinrichs, Heather Trumbower, Robert Kuhn, Donna Karolchik, and the Genome Browser sysadmin team. See the Credits page for a detailed list of the organizations and individuals who contributed to this release.

The UCSC Bioinformatics Group announces 2 seminars and hands-on workshops on the UCSC Genome Browser, presented by OpenHelix, a bioinformatics training, software testing and consulting company.

These introductory sessions are geared towards anyone with a basic knowledge of genomic and biological concepts who is interested in learning how to use the UCSC Genome Browser. No programming experience is required. The seminars will cover the topics necessary to learn how to effectively use the browser tool set, including basic Genome Browser functionality, searching and BLAT use, Table Browser use, creating and using custom annotation tracks, and an introduction to the Gene Sorter. Lectures will be accompanied by hands-on computer exercises conducted directly on the Genome Browser web site.

The first three-hour course will be held on Monday 4 October in Kirkland, WA (Seattle area). Two sessions will be offered: 9:00 a.m. - noon or 1 - 4 p.m.

The second seminar will be held in Cupertino (San Francisco Bay Area) on Tuesday, 5 October. Two sessions will be offered: 1 - 4 p.m. or 6 - 9 p.m.

For registration information, visit the OpenHelix website or call 1-888-861-5051. Academic, student, and early registration discounts are available.

The UCSC Genome Bioinformatics Group has released a Genome Browser and Blat server for the Anopheles gambiae v. MOZ2 draft genome sequence (Feb. 2003). This assembly -- UCSC version anoGam1 -- was produced by the International Anopheles Genome Project and downloaded from Ensembl.

The MOZ2 assembly is a 10x whole genome shotgun assembly. The assembled portion of the genome is about 278 Mbp in length with a total of 8,987 unique scaffolds, the largest scaffold being 23.1 Mbp. Approximately 85% of the sequence has been assigned to chromosomal locations. Chromosome arms chr2L, chr2R, chr3L, chr3R, and chrX are represented by 13, 49, 42, 28, and 10 large scaffolds respectively. No scaffolds have yet been assigned to the Y chromosome. The unassigned scaffolds, concatenated together in arbitrary order, can be found in the artificial unknown "chromosome" chrUn.

For more information about the initial A. gambaie assembly, see Holt et al. (2002), The Genome Sequence of the Malaria Mosquito Anopheles gambiae, Science 2002 298:129-149.

Downloads of the anoGam1 data and annotations may be obtained from the UCSC Genome Browser FTP server or Downloads page. The anoGam1 annotation tracks were generated by UCSC and collaborators worldwide.

Many thanks to the International Anopheles Genome Project and Ensembl for providing the sequence of this genome. The UCSC team who produced this browser are Angie Hinrichs, Galt Barber, Donna Karolchik, and sysadmins Paul Tatarsky and Jorge Garcia. See the Credits page for a detailed list of the organizations and individuals who contributed to this release.

The Zv3 Zebrafish genome assembly (UCSC version danRer1) is now available on the UCSC Genome Browser and Blat server. This assembly was produced by The Wellcome Trust Sanger Institute, Hinxton, UK, in collaboration with the Max Planck Institute for Developmental Biology in Tuebingen, Germany, and the Netherlands Institute for Developmental Biology (Hubrecht Laboratory), Utrecht, The Netherlands.

The Zv3 assembly consists of 1,459,115,486 bp in 58,339 supercontigs, with a sequence coverage of approximately 5.7X. This zebrafish assembly is the first to be tied to the FPC map: 1,083,447,588 bp (74%) of the sequence were mapped in this way. Please note that this is a preliminary assembly; a high level of misassembly is present due to polymorphisms in the DNA source.

For more information about this assembly, see the Sanger Institute's Danio rerio Sequencing Project web page. UCSC plans to release the Zv4 version of the zebrafish assembly on the Genome Browser in Fall '04.

Downloads of the Zebrafish data and annotations can be obtained from the UCSC FTP site or Downloads page. The danRer1 annotation tracks were generated by UCSC and collaborators worldwide. See the Credits page for a detailed list of the organizations and individuals who contributed to the success of this release.

We'd like to thank The Wellcome Trust Sanger Institute and their collaborators for providing this assembly. A special thanks to Yi Zhou, Anthony DiBiase and Leonard Zon from the Children's Hospital in Boston, MA, USA for their collaboration on this release. The UCSC Zebrafish Genome Browser team is Rachel Harte, Heather Trumbower, and Donna Karolchik.

The latest human genome reference sequence (NCBI Build 35, May 2004) is now available as database hg17 in the UCSC Genome Browser and Blat server. This sequence was obtained from NCBI and was produced by the International Human Genome Sequencing Consortium.

Bulk downloads of the data are available via FTP at ftp://hgdownload.cse.ucsc.edu/goldenPath/hg17 or through the Downloads link on this page. We recommend that you use FTP rather than HTML for the download of large or multiple files.

We'd like to thank NCBI and the International Human Genome Sequencing Consortium for furnishing the data, and the UCSC team members who contributed to this release: Hiram Clawson, Terry Furey, Heather Trumbower, Robert Kuhn, Donna Karolchik, Kate Rosenbloom, Angie Hinrichs, Rachel Harte, Jim Kent and our sysadmin team Patrick Gavin, Jorge Garcia, and Paul Tatarsky.

The UCSC Genome Bioinformatics Group has released a Genome Browser and Blat server on a second species of fruitfly, D. yakuba. The April 2004 Release 1.0 of this genome (UCSC version droYak1) was sequenced and assembled by the Genome Sequencing Center, Washington University (WUSTL) School of Medicine in St. Louis.

D. yakuba is closely related to the model organism, D. melanogaster, with which it shared a common ancestor approximately 10 million years ago. The D. yakuba genome is largely alignable to the D. melanogaster genome, but differs sufficiently to offer an interesting study of sequence divergence between the two species. D. yakuba occupies a critical intermediate position among several Drosophila species that will facilitate evolutionary studies among the fruitflies. For information about the D. yakuba assembly and statistics, see the WUSTL Genome Sequencing Center Drosophila yakuba web page.

Downloads of the droYak1 data and annotations can be obtained from the UCSC Genome Browser FTP server or Downloads page. The droYak1 annotation tracks were generated by UCSC and collaborators worldwide.

Thanks to the Genome Sequencing Center at WUSTL School of Medicine for providing the sequence and assembly of this genome. The UCSC D. yakuba Genome Browser was produced by Angie Hinrichs, Michael Chalup, and Donna Karolchik. See the Credits page for a detailed list of the organizations and individuals who contributed to the success of this release.

The latest mouse assembly -- Build 33 from NCBI (UCSC version mm5) -- is now available via the UCSC Genome Browser and Blat server. This assembly includes approximately 2.6 gigabases of sequence. Chromosome 11 is finished in Build 33; the Sanger Institute has provided a corresponding agp file. The whole genome N50 for this assembly is 22.3 Mb, in comparison to 17.7 Mb for the previous build. Please note: the UCSC mm5 assembly contains only the reference strain C57BL/6J.

This assembly is a composite version in which phase 3 High Throughput Genome Sequence (HTGS) was merged with the Mouse Genome Sequencing Consortium v3 Whole Genome Shotgun Assembly (MGSCv3). The assembly was performed by NCBI using a "combined" tiling path that was created automatically for the most part, but was manually curated in places. This facilitated the placement of finished sequence in the context of the MGSCv3 assembly. Draft sequence was not included in this build: the slight increase in coverage gained by using this would have been offset by the increase in build errors.

More information about Build 33 will be available soon in the NCBI assembly notes and Build 33 statistics.

The mm5 sequence and annotation data may be downloaded from the Genome Browser FTP server or Downloads web page. The mm5 annotation tracks were generated by UCSC and collaborators worldwide.

We'd like to thank Deanna Church, Richa Agrawala, and the Mouse Genome Sequencing Consortium for this assembly. We'd also like to acknowledge the work of the UCSC mm5 team: Fan Hsu, Hiram Clawson, Angie Hinrichs, Heather Trumbower, Mark Diekhans, Donna Karolchik and our systems administrators Jorge Garcia, Patrick Gavin and Paul Tatarsky.

The v1.0 C. intestinalis draft assembly from the US DOE Joint Genome Institute is now available for study using the UCSC Genome Browser and Blat server (UCSC database ci1).

The whole genome shotgun assembly was constructed with the JGI assembler (JAZZ) paired-end sequencing reads at a coverage of 8.2X. The draft contains 116.7 million bp of nonrepetitive sequence in 2,501 scaffolds greater than 3 kb. 60 Mbp of this has been assembled into 117 scaffolds longer than 190 Kbp, and 85% of the assembly (104.1 Mbp) is found in 905 scaffolds longer than 20 kb. The assembly, gene modeling and analysis were performed at the JGI.

For more information about the ci1 assembly, see the JGI C. intestinalis project page. Additional information and an analysis of the euchromatic regions of this genome may be found in Dehal et al., The Draft Genome of Ciona intestinalis: Insights into Chordate and Vertebrate Origins. Science. 2002 Dec 13;298(5601):2157-67.

Bulk downloads of the sequence and annotation data are available via the Genome Browser FTP server or Downloads page. The ci1 annotation tracks were generated by UCSC and collaborators worldwide. See the Credits page for a detailed list of the organizations and individuals who contributed to this release.

Many thanks to the JGI and their collaborators for providing the v1.0 sequence and annotations. The ci1 Genome Browser was produced by Brian Raney, Galt Barber, Heather Trumbower, Robert Kuhn, Donna Karolchik and the Genome Browser sysadmin team - Patrick Gavin, Jorge Garcia, and Paul Tatarsky. We'd also like to thank Tom Pringle for his technical input and Mark Diekhans for his work on the incremental updates for this release.

UCSC has released a Genome Browser and Blat server on the July 2004 v1.0 dog genome sequenced and assembled by the Broad Institute of MIT and Harvard and Agencourt Bioscience. The whole genome shotgun (WGS) sequence is based on 7.6X coverage of the dog genome, assuming a WGS assembly size of 2.4 Gb. The assembly has an N50 contig length of 123 kb and an N50 supercontig length of 41.6 Mb.

The dog genome, which contains approximately 2.5 billion base pairs, is similar in size to the genomes of humans and other mammals. The boxer breed was selected for the initial sequencing effort, based on the lower variation rate in its genome relative to other breeds. In addition to the boxer, samples from nine other dog breeds, four wolves and a coyote are being used to generate an initial set of single nucleotide polymorphisms (SNPs) to facilitate disease studies. The SNPs should be available soon from dbSNP.

For more information about the dog draft assembly, see the NHGRI press release.

The dog sequence and annotation data can be downloaded from the UCSC Genome Browser FTP server or downloads page. These data have specific conditions for use.

Many thanks to the Broad Institute of MIT and Harvard, NHGRI, Agencourt Bioscience, Children's Hospital Oakland Research Institute, Centre National de la Recherche Scientifique, North Carolina State University, and Fred Hutchinson Cancer Research Center for their contributions to the sequencing, assembly, and mapping efforts. The initial canFam1 annotation track set, generated by the UCSC Genome Bioinformatics Group, will soon be supplemented by annotations from collaborators worldwide. See the credits page for a detailed list of the organizations and individuals who contributed to the success of this release.

We'd like to announce the release of UCSC Genome Browser features tailored to the ENCODE project community, including an ENCODE-specific page to highlight the ENCODE contributors and their work, guidelines for data submission, and a list of specific links to ENCODE regions in the Genome Browser.

The initial resources include sequences for the current human assemblies (hg16, hg15, hg13, and hg12), sequence of the comparative species from NISC, tools for coordinate conversion between human assemblies, format descriptions for data submission, and contact information for help with submitting annotation data and analyses. Bulk downloads of the sequence and annotations may be obtained from the ENCODE Project Downloads page.

We'd like to thank NHGRI for their support of this project and the various contributors of annotations and analyses.

The UCSC Bioinformatics Group announces a seminar and hands-on workshop on the UCSC Genome Browser, presented by OpenHelix, a bioinformatics training, software testing and consulting company.

This introductory session is geared towards anyone with a basic knowledge of genomic and biological concepts who is interested in learning how to use the UCSC Genome Browser. No programming experience is required. The seminar will cover the topics necessary to learn how to effectively use the browser tool set, including basic Genome Browser functionality, searching and BLAT use, Table Browser use, creating and using custom annotation tracks, and an introduction to the Gene Sorter. The lecture will be accompanied by hands-on computer exercises conducted directly on the Genome Browser web site.

The three-hour course will be held at Tufts University School of Medicine, 145 Harrison Street, Boston, MA, on Tuesday, August 10th. Two sessions will be offered: 1 - 4 p.m. or 6 - 9 p.m. For registration information, visit the OpenHelix website or call 1-888-861-5051. Academic, student, and early registration discounts are available.

What's in a name? In an effort to clarify the role of the UCSC Family Browser, we have changed its name to the UCSC Gene Sorter. We think this name better describes this tool, which lets the user collect information on groups of genes that may be related in many different ways. The Gene Sorter provides a wealth of information on gene expression, protein homology (both within and across species), GO terms, and Pfam domains, cross links to many other databases, and much more.

If you haven't already tried this tool, we encourage you to give it a spin. You'll find it at http://genome.ucsc.edu/cgi-bin/hgNear, or click the "Gene Sorter" link on any Genome Browser menu bar.

As a follow-up to last week's FTP site switch, we are changing the location of the UCSC Genome Browser downloads site to http://hgdownload.cse.ucsc.edu/. All downloadable files currently located in http://genome.ucsc.edu/goldenPath will be moved to the new server. Please make a note of the new URL and update any references to it. Users accessing downloads through the Genome Browser Downloads page will be redirected automatically to the new location.

We have changed the URL for the UCSC Genome Browser ftp site to ftp://hgdownload.cse.ucsc.edu/. This replaces the old URL of ftp://genome.ucsc.edu/. The old URL will be disabled within a few days. Please make a note of the new URL and update any references to it.

UCSC has released a Genome Browser and Blat server on an updated version of the C. elegans genome. The March 2004 assembly -- UCSC version ce2 -- is based on sequence version WS120 deposited into WormBase as of 1 March 2004. This assembly has a finishing error rate of 1:10,000.

The ce2 sequence and annotation data may be downloaded from the Genome Browser FTP server or Downloads web page. The ce2 annotation tracks were generated by UCSC and collaborators worldwide.

We'd like to thank the Genome Sequencing Center at Washington University in St. Louis and the Sanger Institute for their collaborative work in sequencing the C. elegans genome. Many thanks to the WormBase consortium for making the worm sequence publicly available. We'd also like to acknowledge the UCSC team who contributed to this release: Rachel Harte (lead engineer), Hiram Clawson (WABA and miRNA annotations), Mike Chalup (QA), Galt Barber (QA), Heather Trumbower (QA), and Donna Karolchik (documentation).

Proteome Browser functionality is now available on the Oct. 2003 mouse genome assembly (mm4). Protein information may be viewed for most genes in the Known Genes track by clicking the Proteome Browser link on the gene's details page. For more information on the UCSC Proteome Browser, see the news release dated 10 March 2004.

In conjunction with this release, the Known Genes and Gene Family Browser protein data have been updated to the Swiss-Prot version dated 15 March 2004.

The UCSC Bioinformatics Group announces a seminar and hands-on workshop on the UCSC Genome Browser, presented by OpenHelix, a bioinformatics training, software testing and consulting company.

This introductory session is geared towards industry and academic biologists engaged in genomics research. No programming experience is required. The seminar will cover the topics necessary to learn how to effectively use the browser tool set, including basic Genome Browser functionality, searching and BLAT use, Table Browser use, creating and using Custom Tracks, and an introduction to the Family Browser. The lecture will be accompanied by hands-on computer exercises conducted directly on the Genome Browser web site.

The three-hour course will be held at the UCSC Extension Campus computer lab in Cupertino, CA on Thursday, May 6th 6-9 p.m. For pricing information or to reserve a seat in the class, visit the OpenHelix web site or call 1-888-861-5051. Pre-registration is required. Academic, student, and early registration discounts are available.

Expression data from the GNF Gene Expression Atlas 2 are now available on the July 2003 human genome assembly on the UCSC web site. The data may be viewed graphically in the Family Browser or via the GNF Atlas 2 track in the Genome Browser. The track data contain 2 replicates each of 61 mouse tissues and 79 human tissues run over Affymetrix microarrays.

We'd like to thank the Genomics Institute of the Novartis Research Foundation (GNF) for providing the expression data underlying the browser displays. More information on the data will be available in the paper Su et al. "A gene atlas of the mouse and human protein-encoding transcriptomes" (in press - PNAS).

We are proud to announce a new addition to the UCSC family of genome browsing and analysis tools. The UCSC Proteome Browser presents a rich set of useful protein properties as well as links to several protein and genomic data sources on the Web. For the first time, Genome Browser users can have both the genome and proteome worlds at their fingertips simultaneously. The browser is accessible from the Genome Browser via the "Proteome Browser" link on the details page of any gene in the Known Genes track. The initial release is available only on Human Build 34 (hg16); Proteome Browsers for the latest mouse and rat assemblies will follow.

For each protein, the browser displays the corresponding genomic exon structure and its amino acid sequence. Several protein property tracks are aligned to the sequence to help a user pinpoint regions of interest. Additional properties are plotted with histograms against genome-wide protein data to highlight significant trends and anomalies.

The Proteome Browser is tightly coupled with the UCSC Genome Browser and UCSC Gene Family Browser, allowing easy navigation among the tools. For example, clicking on an exon in the Proteome Browser tracks display brings up the Genome Browser tracks page showing the genomic region of the exon together with a wealth of relevant data. Similarly, clicking on the Proteome Browser's "Family Browser" link displays related gene family information.

The v1.0 release of the browser offers a variety of data tracks, including amino acid and DNA sequence, exon boundaries, hydrophobicity, polarity, cysteine and predicted glycosylation sites, Superfamily/SCOP domains, and amino acid anomalies. In addition, the browser includes histograms of several properties on a genome-wide scale: pI, molecular weight, exon count, number of cysteines, InterPro domain counts, hydrophobicity, amino acid frequencies and anomalies.

The Proteome Browser also provides links to a variety of external sites containing supplementary information on the protein, including SwissProt, InterPro and Pfam domains, 3-D structures at PDB and UCSF ModBase, and pathway maps of KEGG, BioCarta (CGAP), and BioCyc.

We'd like to thank SwissProt for sharing their high quality protein data and the pI calculation algorithm, as well as the other external data sites linked to by the Proteome Browser. We'd also like to acknowledge the hard work of Fan Hsu, lead engineer on the project, and Jim Kent, Tom Pringle, Donna Karolchik, and Robert Kuhn. The project received technical input, review and support from several other members of the UCSC Bioinformatics group.

We've added the chicken genome to the collection of assemblies available in the UCSC Genome Browser and Blat Server. The Feb. 2004 assembly (UCSC version galGal2) was produced by the Genome Sequencing Center at the Washington University School of Medicine in St. Louis. The source of this sequence was a female inbred Red Jungle Fowl (Gallus gallus), the ancestor of domestic chickens. The chicken genome is the first of the avian genomes to be sequenced.

The genome has been sequenced to 6.63X coverage. Approximately 88% of the sequence has been anchored to chromosomes, which include autosomes 1-24, 26-28, and 32, and sex chromosomes W and Z. (In contrast to mammals, the female chicken is heterogametic (ZW) and the male is homogametic (ZZ).) The remaining unanchored contigs have been concatenated into the virtual chromosome "chrUn", separated by gaps of 10,000 bp. The chicken mitochondrial sequence is also available as the virtual chromosome "chrM".

Washington University School of Medicine in St. Louis created the physical map for this release. Genetic mapping and linkage analysis were produced through a collaborative effort led by Martien Groenen at Wageningen University in the Netherlands. SNP data based on three strains of domestic chickens will soon be available in GenBank from an international team of scientists led by the Beijing Genomics Institute in China and supported by the Wellcome Trust in Britain.

The chicken is considered to be the premier non-mammalian vertebrate model organism. It plays an important role in the research of viruses and cancer, and is a primary model for the study of embryology and development. From an evolutionary standpoint, the chicken's position provides a good intermediate data point between mouse and fugu. Comparative genomics analyses between the chicken and other sequenced organisms should yield valuable information on the evolution of gene order and arrangement, thus improving our understanding of the structure and function of genes.

To facilitate comparative genomics studies, alignments of the chicken sequence to the human genome will be available in the Genome Browser later this week. Downloads of the comparative data are currently available through the Downloads page (see below).

For more information about the release of the chicken genome assembly, see the NHGRI press release. Additional background on the rationale behind the chicken genome sequencing effort can be found in the sequencing proposal.

Bulk downloads of the chicken sequence and annotations may be obtained from the Genome Browser FTP server or Downloads page. These data have specific conditions for use.

We'd like to thank the Genome Sequencing Center at the Washington University School of Medicine in St. Louis, Wageningen University, and the Chicken Mapping Consortium for providing these data. The chicken browser annotation tracks were generated by UCSC and collaborators worldwide. See the Credits page for a detailed list of acknowledgements. The UCSC Chicken Genome Browser was produced by Angie Hinrichs, Heather Trumbower, Rachel Harte, and Donna Karolchik.

We are happy to announce the release of a Genome Browser and Blat server for the chimpanzee (Pan troglodytes). The 13 Nov. 2003 Arachne assembly -- labeled Chimp Build 1 Version 1 (UCSC version panTro1) -- was produced by the Chimpanzee Genome Sequencing Consortium.

This assembly covers about 95 percent of the genome and is based on 4X sequence coverage. It is composed of 361,782 contigs with an N50 length of 15.7 kb, and 37,849 supercontigs having an N50 length of 8.6 Mb (not including gaps). The total contig length is 2.73 Gb, spanning 3.02 Gb. The DNA donor for this genome assembly, "Clint", is a captive-born West African chimpanzee (Pan troglodytes verus). Background information on the chimp genome sequencing project and the initial news release about the chimp assembly can be found on the NHGRI website.

Research has indicated that the human and chimp genomes probably differ by approximately one percent. Because of this close relationship between chimpanzees and humans, the assembly should facilitate comparative analyses of the two genomes that have not been possible with other species that have been sequenced to date.

The initial release of the Chimp Browser provides several annotation tracks comparing the chimp and human genomes. More comparative annotations will be added in upcoming weeks.

Bulk downloads of the chimp sequence and annotations may be obtained from the Genome Browser FTP server or Downloads page. The complete set of sequence reads is available at the NCBI trace archive. Blat searches on chrUn_random are not supported in the initial release, but will be available soon.

We'd like to thank NHGRI, the Eli & Edythe L. Broad Institute at MIT/Harvard, and Washington University at St. Louis School of Medicine for providing this sequence, and LaDeana Hillier, Washington University School of Medicine, and the Broad Institute for their work on the alignments. The chimpanzee browser annotation tracks were generated by UCSC and collaborators worldwide.

The UCSC team who worked on this release consisted of Kate Rosenbloom, Jim Kent, Hiram Clawson, Heather Trumbower, Robert Kuhn, Donna Karolchik, and the Genome Browser sysadmin team.

The Genome Browser project now has a UCSC-supported mirror site that may be used during power or network outages on the UCSC campus. The mirror -- which can be found at http://genome.brc.mcw.edu/ -- is located at the Medical College of Wisconsin in Milwaukee. The site will be updated regularly by UCSC with the latest data and software to closely replicate the main Genome Browser site at http://genome.ucsc.edu. Please continue to use the UCSC-based site for routine Genome Browser and Blat access.

We'd like to thank the Department of Physiology at the Medical College of Wisconsin -- and in particular Jeff Nie and Greg McQuestion -- for their resources and collaboration on this project. We'd also like to acknowledge the hard work of UCSC's Paul Tatarsky, who invested many hours in arranging the collaboration and setting up the mirror.

We've discovered a handful of hg16 chrN_random_gap and chrN_random_gold tables on our public server that are out of date. We have replaced the following tables with updated versions:

• chr4_random_gap
• chr4_random_gold
• chr8_random_gap
• chr8_random_gold
• chrX_random_gap
• chrX_random_gold
• chrUn_random_gap
• chrUn_random_gold

Many thanks to Grigoriy Kryukov for discovering this problem. We apologize for any inconvenience this may have caused to our users.

We are proud to add yeast (S. cerevisiae) to our growing list of genome assemblies. The study of brewer's yeast, the most basic eukaryotic model system, has led to important discoveries in a wide variety of areas, including metabolism, centromeres, recombination, cell division control, meiosis and splicing.

This assembly (UCSC version sacCer1) is based on sequence dated 1 Oct. 2003 in the Saccharomyces Genome Database (SGD). The sequence, open reading frame (ORF), and gene annotations were downloaded from the site ftp://genome-ftp.stanford.edu/pub/yeast/data_download.

The S288C strain was used in this sequencing project. Reference information for each chromosome may be found in the SGD Systematic Sequencing Table. For more information about the yeast genetic and physical maps, see the paper Cherry JM et al. Genetic and physical maps of Saccharomyces cerevisiae. Nature 1997 387(6632 Suppl):67-73.

Downloads of the yeast data and annotations may be obtained from the UCSC Genome Browser FTP server or Downloads page.

We'd like to thank Stanford University, the SGD, the University of California San Francisco (UCSF), Washington University in St. Louis, and the Eli & Edythe L. Broad Institute at MIT/Harvard for providing the data and annotations for this assembly. We'd also like to acknowledge the UCSC team who worked on this release: Jim Kent, Heather Trumbower, Robert Kuhn, Donna Karolchik, and our sysadmin team.

UCSC has released alignments of the Nov. 2003 chimpanzee draft assembly to the July 2003 human assembly in the Genome Browser. These alignments may be viewed on the Human July 2003 assembly. This release coincides with today's announcement by the National Human Genome Research Institute (NHGRI) of the first draft assembly of the chimpanzee genome.

The set of human/chimpanzee alignments consists of a reciprocal best-in-genome net track and a chimp chain track. These alignments were generated using the blastz program developed at Pennsylvania State University and the programs blat, axtChain, chainNet, and netSyntenic developed at UCSC by Jim Kent. Research scientists should find these tracks useful for locating orthologous regions and studying genome rearrangement in the two species.

For more information about the alignment tracks, refer to the track description pages. The tables may be downloaded from the Genome Browser FTP server's hg16 database directory. The chimp sequence and alignment data are downloadable from the hg16 human/chimp alignments directory.

The chimp sequence used in these alignments was obtained from the 13 Nov. 2003 Arachne assembly. We'd like to thank NHGRI, the Eli & Edythe L. Broad Institute at MIT/Harvard, and Washington University School of Medicine for providing this sequence, and LaDeana Hillier, Washington University School of Medicine, and the Whitehead Institute for their work on the alignments. We'd also like to acknowledge the members of the UCSC team who contributed to the release of these alignments in the Genome Browser: Jim Kent, Kate Rosenbloom, Heather Trumbower, and Donna Karolchik.

We have released a Genome Browser and Blat server for the latest mouse genome assembly, NCBI Build 32 (UCSC v. mm4). Build 32 is a composite assembly in which chromosomes were assembled by two slightly different algorithms depending on the available mapping data. Chromosomes 2, 4, 5, 7, 11, 15, 18, 19, X, and Y were assembled using a clone-based tiling path file (TPF) provided by the Mouse Genome Sequencing Consortium (MGSC), with whole genome shotgun sequence used to fill gaps when necessary. The remaining chromosomes were assembled using the MGSCv3 whole genome shotgun assembly as the TPF and merging High Throughput Genomic Sequence (HTGS) as needed. The UCSC mm4 assembly contains only the reference strain C57BL/6J.

Build 32 includes 2.6 gigabases of sequence, 1.2 Gb of which is finished. We estimate that 90-96 percent of the mouse genome is present in the assembly. For more information about this version, see the NCBI assembly notes and Build 32 statistics.

The mm4 sequence and annotation data may be downloaded from the UCSC Genome Browser FTP site or downloads page.

We'd like to thank the Deanna Church, Richa Agrawala, and the Mouse Genome Sequencing Consortium for this assembly. We'd also like to acknowledge the work of the UCSC mm4 team: Hiram Clawson (lead), Terry Furey, Kate Rosenbloom, Heather Trumbower, Bob Kuhn and Donna Karolchik, and our systems administrators Patrick Gavin, Jorge Garcia and Paul Tatarsky.

The fruitfly, one of the first organisms to be used in systematic scientific investigations, has been the subject of intensive study in genetics for nearly a century and remains a major model organism in biomedical research, population biology and evolution. We are pleased to add the fruitfly to the roster of assemblies available on our site.

Downloads of the Drosophila data and annotations may be obtained from the UCSC Genome Browser ftp site.

We'd like to thank BDGP and the Flybase Consortium (Harvard University, University of Cambridge, Indiana University, the University of California, Berkeley and the European Bioinformatics Institute (EBI)) for providing the sequence, assembly, and analysis of this genome. We'd also like to acknowledge the members of the UCSC Genome Bioinformatics group who contributed to this release: Angie Hinrichs (lead engineer), Heather Trumbower, Robert Kuhn, Donna Karolchik, and Jim Kent and the system administrators Jorge Garcia, Patrick Gavin and Paul Tatarsky.

Previously, these tables were updated only when we loaded a new genome assembly into the Genome Browser or made a major revision to a table. By updating the data on a nightly basis, we are able to provide researchers with the most current versions available in GenBank. All new genome assemblies released after this date will incorporate the incremental update technology.

Data are updated on the following schedule:
- native and xeno mRNA and refSeq tracks - updated daily at 4:30 p.m. Pacific Time (weekdays), early Saturday morning (weekends)
- EST data - updated once per week on Saturday morning
- downloadable data files - updated weekly on Sunday morning
- outdated sequences - removed once per quarter

Mirror sites are not required to migrate to an incremental update process, and should not experience problems as a result of this upgrade. Mirror site questions should be addressed to genome-mirror@soe. ucsc.edu.

We'd like to acknowledge the hard work of Mark Diekhans in the implementation of this new feature, and thank the QA and sysadmin teams (particularly Paul Tatarsky) for their support in this release.

Many of our users have requested a batch query utility that will allow them to paste in or upload a list of terms on which to search. You can now do this by clicking the "Item name/accession" button, then uploading a list of search terms by selecting the "Paste in" or "Upload" option. Note that the Paste option supports wildcards, but the Upload option does not.

We've also added 2 new lists of searchable tables/tracks. The Browser tracks list contains the names of annotation tracks in the currently selected assembly as they appear in the Genome Browser. This list is useful if you don't know the name of the underlying database table that contains the data in which you're interested. The Custom tracks list contains the names of all custom annotation tables currently loaded into the Genome Browser for the given assembly. This includes tracks that have been created/loaded by the user as well as custom annotations created via the Table Browser.

If you have feedback or questions about the Table Browser, please send us email at genome@soe.ucsc.edu.

To access the Family Browser, click the link on the top menu on this page. The first time you use the Browser, the application will display a brief overview of the tool and information for starting and configuring it. To read a more detailed description of the Browser, see the Family Browser User's Guide.

We are always interested in hearing feedback about the tools on our site. If you have comments or questions about the Family Browser, please email us at genome@soe.ucsc.edu.

We have added the Takifugu rubripes (Japanese pufferfish) assembly to the growing collection of genomes available in the UCSC Genome Browser and Blat servers. The Fugu v.3.0 (Aug. 2002) whole genome shotgun assembly -- which is the fourth vertebrate assembly to be added to the UCSC Genome Browser -- was provided by the US DOE Joint Genome Institute (JGI) as part of the International Fugu Genome Consortium led by JGI and the Singapore Institute of Molecular and Cell Biology (IMCB).

This assembly was constructed with the JGI assembler, JAZZ, from paired end sequencing reads produced by JGI and IMCB, at JGI, Myriad Genetics, and Celera Genomics, resulting in a sequence coverage of 5.7X. All reads are plasmid, cosmid, or BAC end sequences, with the predominant coverage derived from 2 Kb insert plasmids. This assembly contains 20,379 scaffolds totaling 319 million base pairs. The largest 679 scaffolds total 160 million base pairs. The Fugu genome was annotated using the Ensembl system by the Fugu informatics group at IMCB.

The Fugu, which was one of the first vertebrate genomes to be draft-sequenced after human, serves an important role in the exploration of the human genome. In contrast to other vertebrates that have been sequenced, the intergenic and intron regions of the Fugu are highly compressed and uncluttered with repetitive sequence, resulting in a genome that is unusually compact in size. The Fugu genome has proved useful in gene discovery and the identification and characterization of gene regulatory elements in other genomes.

Bulk downloads of the Fugu sequence and annotation data are available via FTP at ftp://hgdownload.cse.ucsc.edu/goldenPath/fr1 or through the Downloads link on the Genome Browser home page. We recommend that FTP be used rather than HTML for the download of large or multiple files.

We'd like to thank JGI and the other members of the International Fugu Genome Consortium, including IMCB, the UK Human Genome Mapping Project (Hinxton), the Molecular Sciences Institute (Berkeley) and the Institute for Systems Biology (Seattle), for providing the sequence, assembly, and analysis of this genome. We'd also like to acknowledge the members of the UCSC Genome Bioinformatics group who contributed to this release: Kate Rosenbloom (lead engineer), Heather Trumbower, Robert Kuhn, Donna Karolchik, and Jim Kent.

The NISC Comparative Sequencing Program data may be accessed by clicking the Browser link on the Genome Browser home page and then selecting the "Zoo" option from the genome list.

The research team, led by NHGRI Scientific Director Eric D. Green, included scientists from Pennsylvania State University, University of California Santa Cruz (UCSC), and the University of Washington in Seattle. In the study, the investigators systematically compared the patterns of transposon insertions among the species' sequences. One key result of the analysis was the confirmation of recently proposed mammalian evolutionary trees suggesting that primates are more closely related to rodents than to carnivores or artiodactyls. Another significant outcome was the discovery of a substantial number of previously unidentified non-coding DNA segments that are conserved across a wide range of species. Many of these regions could be identified only through comparisons of sequence from multiple species, demonstrating the importance of studying the genomes of a wide range of organisms as a means for identifying functional elements in the human genome.

UCSC built a customized version of the browser to display the target region for this study, allowing scientists to interactively explore the data and predictions generated by this project, contribute data of their own, and track the project as data from additional species are generated. In addition to the browser, the UCSC team also contributed to the analytical portion of the study. Mathieu Blanchette identified the regions that are most highly conserved among species. Adam Siepel performed the phylogenetic analysis of rates of substitution. The UCSC team worked with Arian Smit to obtain definitive evidence that rodents branched off from the common ancestor later than carnivores and artiodactyls.

For more information on the NISC study, see the Science Daily press release. Flat files of the assembled sequence and annotations may be obtained from http://www.nisc.nih.gov/data/ or via the Downloads link on the Genome Browser home page.

We'd like to thank the NISC Comparative Sequencing Program team for providing the data and comparative analysis for this Genome Browser release. Special thanks go to Elliott Margulies at NHGRI for serving as the main liaison between NHGRI and UCSC, and for contributing several annotation tracks to the browser. We'd also like to acknowledge the efforts of the many faculty, grad students, and staff members of the UCSC Genome Bioinformatics group who contributed to the research effort and browser development for this project.

There are 2,843,433,602 finished sequenced bases in the ordered and oriented portion of the assembly, which is an increase of 0.4 percent, or approximately 11 Mb, over the Build 33 assembly. Of particular note in this release is the addition of the pseudoautosomal regions of the Y chromosome. This sequence was taken from the corresponding regions in the X chromosome and is an exact duplication of that sequence.

Some sequence joins between adjacent clones in this assembly could not be computationally validated because the clones originated from different haplotypes and contained polymorphisms in the overlapping sequence, or the overlap was too small to be to be reliable. In these instances, the sequencing center responsible for the particular chromosome has provided data to support the join in the form of an electronic certificate. The Build 34 certificates may be reviewed here.

Bulk downloads of the data are available via FTP at ftp://hgdownload.cse.ucsc.edu/goldenPath/hg16 or through the Downloads link. We recommend that FTP be used rather than HTML for the download of large or multiple files.

We'd like to thank NCBI and the International Human Genome Sequencing Consortium for furnishing the data, and the UCSC team members who contributed to this release: Terry Furey, Hiram Clawson, Heather Trumbower, Mark Diekhans, Robert Baertsch, Donna Karolchik, Jim Kent and our sysadmin team Patrick Gavin, Jorge Garcia, and Paul Tatarsky.

The 3.x assemblies reflect several sequence additions and software improvements over the previous 2.x assemblies, including the sequencing of over 1100 new BACs to cover gaps, an improved marker set from the Medical College of Wisconsin, a new FPC map from the BC Genome Sciences Centre, and improved linking of bactigs. For detailed information and statistics about the 3.x assemblies, see the Baylor HGSC README.

Downloads of the rat sequence and annotation data are available at ftp://hgdownload.cse.ucsc.edu/goldenPath/rnJun2003/ or via the Downloads link on this page. These data are made available with specific conditions for use.

We'd like to thank the Rat Genome Sequencing Consortium and Baylor HGSC for providing this assembly, collaborators from other institutions who have contributed annotations, and Arian Smit for updating RepeatMasker for this release. We'd also like to acknowledge the contributions of several individuals at UCSC, including Hiram Clawson, Heather Trumbower, Robert Kuhn, Yontao Lu, Terry Furey, Mark Diekhans, Robert Baertsch, Donna Karolchik, Jim Kent, and our sysadmin team Jorge Garcia, Patrick Gavin, and Paul Tatarsky.

We are pleased to add the nematodes to the roster of genomes available on our site. C. elegans is a major model organism used for biomedical research, and is the first multicellular animal to have a fully sequenced genome. In contrast, the whole genome shotgun assembly of the C. briggsae genome is estimated to have achieved 98% coverage. Draft chromosome sequences are not available for C. briggsae, due to the lack of dense chromosomal maps that allow assignment of ultracontigs to chromosomal locations. As a result, all data in the C. briggsae browser maps to chrUn.

Both worms played a significant role in the early history of the UCSC Genome Browser. The browser code originated with a C script that displayed a splicing diagram for a gene prediction from C. elegans. Tracks for mRNA alignments and for homology with C. briggsae were added, and the tool morphed into the precursor of the Genome Browser, the "Intronerator" (Kent, WJ and Zahler, AM (2000). The intronerator: Exploring introns and alternative splicing in C. elegans. Nucleic Acids Res. 28: 91-93).

Downloads of the C. elegans sequence and annotation data are available at ftp://hgdownload.cse.ucsc.edu/goldenPath/ceMay2003/; C. briggsae downloads can be found at ftp://hgdownload.cse.ucsc.edu/goldenPath/cbJul2002/. Both genomes can also be downloaded via the Downloads link on this page.

We'd like to thank the Genome Sequencing Center at Washington University in St. Louis and the Sanger Institute for their collaborative work in sequencing the C. elegans and C. briggsae genomes. Many thanks to the WormBase consortium for making the worm sequence publicly available. We'd also like to acknowledge several UCSC people who contributed to this release: Hiram Clawson (browser and annotation tracks engineering), Jim Kent (WABA and chaining/netting), Al Zahler (WABA), Heather Trumbower (QA and project management), and Donna Karolchik (project management and documentation).

This browser marks a departure from our usual collection of vertebrate genomes. Its inception was inspired by one of our engineers - Angie Hinrichs - who was vacationing in New Zealand when the SARs draft assembly was initially released. Struck by the impact of SARS in that part of the world, she downloaded the sequence and built the initial tracks from a terminal at an Internet cafe! The rest of the team joined in on the grassroots effort, generating the additional annotations and SAM T02 protein analyses and predictions. Victor Solovyev chimed in with Fgenesv+ gene predictions from Softberry Inc. UCSC does not intend to provide a comprehensive collection of viral genomes in the future, but will maintain this browser as long as scientific and public interest in SARS persists.

Downloads of the annotation data are available at ftp://hgdownload.cse.ucsc.edu/goldenPath/scApr2003/database or via the Downloads link on this page.

We'd like to thank everyone who worked on this release, including Angie Hinrichs, Robert Baertsch, Fan Hsu, Matt Schwartz, Heather Trumbower, Jim Kent, Kevin Karplus, Donna Karolchik, Brian Raney, Hiram Clawson, Kate Rosenbloom, Victor Solovyev, and our extremely dedicated systems administrators Paul Tatarsky, Patrick Gavin, and Jorge Garcia.

In conjunction with this announcement, the UCSC Genome Bioinformatics group is proud to release a genome browser and BLAT server on the reference sequence (NCBI Build 33), along with bulk downloads of the sequence and annotation data. The initial browser provides a preliminary set of annotations that will be expanded in coming weeks. Bulk downloads of the data are available via FTP at ftp://hgdownload.cse.ucsc.edu/goldenPath/10april2003 or through the Downloads link on this page. We recommend that FTP be used rather than HTML for the download of large or multiple files.

The reference sequence covers about 99 percent of the human genome's gene-containing regions, and has been sequenced to an accuracy of 99.99 percent. The missing portions are essentially contained in less than 400 defined gaps that represent DNA regions with unusual structures that can't be reliably sequenced using current technology. The average DNA letter now lies within a stretch of approximately 27,332,000 base pairs of uninterrupted sequence!

Chromosomal sequences for this release were assembled by the International Human Genome Sequencing Consortium sequencing centers and verified by NCBI and UCSC. In some cases, sequence joins between adjacent clones could not be computationally validated, e.g. due to polymorphisms in the overlapping sequence. In situations like this, supporting evidence for the join has been provided by the sequencing center responsible for that particular chromosome. The Non-standard Join Certificates table displays this information. The annotations on the UCSC website have been provided by UCSC and collaborators worldwide. See the Credits page for a detailed list of organizations and individuals who contributed to the success of this release.

We'd like to congratulate the many people worldwide who have worked on the Human Genome Project for this landmark achievement. We'd also like to acknowledge the UCSC Genome Browser project team who worked many long hours to ensure that the genome browser and sequence data were released on time for this announcement: David Haussler, Jim Kent, Terry Furey, Matt Schwartz, Heather Trumbower, Angie Hinrichs, Fan Hsu, Donna Karolchik, Jorge Garcia, Patrick Gavin, Chuck Sugnet, Yontao Lu, Mark Diekhans, Ryan Weber, Robert Baertsch, Krishna Roskin, and the many other students in the UCSC Genome Bioinformatics group.

This update corrects duplications that were assembly artifacts in the previous version and improves the linking of bactigs to create larger "ultrabactigs". Compared with the previous rat assembly, sequence mapped to specific chromosomal coordinates is reduced by about 1.6 percent. Loosely mapped and unmapped sequence is reduced by 17 percent. For more details and statistics on the Jan 2003 assembly, see the Baylor HGSC README for this release.

UCSC has released a Genome Browser and BLAT server for this assembly update. The initial browser contains 16 annotation tracks, with more to follow in coming weeks. Sequence downloads are currently available at ftp://hgdownload.cse.ucsc.edu/goldenPath/rnJan2003/ or via the Downloads link on this page. A complete set of database downloads will be available at the beginning of next week. This data contains specific conditions for use.

Thanks to the Atlas group at Baylor HGSC, the Rat Genome Sequencing Consortium, the UCSC Genome Bioinformatics group, and contributors worldwide for making this release available.

UCSC has released a Genome Browser and BLAT server for the Feb. 2003 Mouse genome. The initial browser contains 14 annotation tracks, with more to follow in coming weeks. Sequence downloads are currently available at ftp://hgdownload.cse.ucsc.edu/goldenPath/mmFeb2003/ or via the Downloads link on this page. Database downloads will be available at the beginning of next week.

Thanks to everybody at UCSC and around the world that contributed to the success of this release!

New functionality in v.17:
-- Numerous enhancements to the table browser that allow the user to conduct more complex and specific searches. New features include support for intersections of tracks, a new summary statistics output format, and the ability to output query results as a custom annotation track that can be viewed in the Genome Browser. The new Table Browser User's Guide contains a detailed description of the new features and provides a wealth of information and examples for conducting various types of searches on the database tables.

-- Two new display modes available for most annotation tracks: pack and squish modes. In pack mode display, annotation track features are f ully displayed, but more than one feature may be displayed on the same line. This greatly reduces the amount of display space needed by a track when a user wishes to view a large number of individual features at one time. Squish mode is similar to pack mode, but displays features at 50% height and without labels. This mode is particularly useful for viewing tracks in which a large number of features align to the same section of a chromosome, e.g. EST tracks.

-- Functional groupings of annotation track controls. This makes it much easier to find a particular item in the track control list and gives a better visual overview of the annotations available in a particular category, e.g. comparative genomics tracks or gene prediction tracks.

-- A mechanism for saving the annotation tracks image in postscript or PDF format. This much requested feature enables Genome Browser users to print an image at high resolution, edit it with a drawing program, or display it in a postscript or PDF viewer.

-- A collection of custom annotation tracks supplied by Genome Browser users and members of the UCSC Genome Bioinformatics lab. Additional contributions to this collection are welcome! Contact genome@soe.ucsc.edu if you have an annotation you'd like to share.

New annotation tracks in v.17:
-- a new Known Genes track (Human Builds 30 & 31, Feb 2002 Mouse, Nov 2002 Rat) that shows known protein coding genes based on proteins from SWISS-PROT, TrEMBL, and TrEMBL-NEW and their corresponding mRNAs from Genbank. Features within the track are color-coded according to origin and review status.

-- a new Superfamily track (Human Build 30, Feb 2002 Mouse) that shows proteins having homologs with known structures or functions. Each entry in the track shows the coding region of a gene (based on Ensembl gene prediction). The feature label consists of the names of all known protein domains coded by the gene, and usually contains structural and/or function descriptions that provide valuable information for getting a quick grasp of the biological significance for the gene.

We have also released several additional annotation tracks on the latest human and rat assemblies in the past month.

Bug fixes in v.17:
-- Approximately 40 bugs (mostly minor problems) have been fixed in this version.

To take full advantage of the new display features in this release, we recommend that you reset your browser to the new default settings. NOTE: you may not want to reset your browser if you have customized settings that you wish to preserve. You can reset your browser by clicking the "Click here to reset" link on the Browser Gateway page.

We hope this new release facilitates your work with the UCSC Genome Browser. If you have any questions or comments about the new release, send email to genome@soe.ucsc.edu.

We're pleased to announce the release of the latest human genome assembly, Build 31 (UCSC version hg13). This assembly was produced at NCBI based on sequence information submitted into GenBank as of Nov. 14, 2002. Release notes for this assembly are available from the NCBI web site. Because UCSC now obtains its assembly directly from NCBI, the UCSC Build 31 data is identical to that of NCBI and Ensembl.

Build 31 is an excellent high-quality assembly that shows a remarkable amount of progress toward the milestone of finishing the human genome. Greater than 95% of the euchromatic region of the genome is now complete, with more than 90% of the sequence in a finished state. The number of clone contig gaps has decreased by one third from the previous assembly, and the overall number of sequenced contigs has been reduced by one half. Seven chromosomes are considered to be in a finished state: 6, 7, 13, 20, 21, 22, and Y.

The initial release of the Build 31 Genome Browser contains 25 annotation tracks, with several more to follow in the upcoming weeks. Bulk downloads of the data are available from our FTP site at ftp://hgdownload.cse.ucsc.edu/goldenPath/14nov2002 or via the Downloads link on this page.

UCSC has generated a set of high-level comparisons of the Build 31 draft sequence against various types of information (STS maps, BAC end pairs, and clone overlaps). This information, as well as statistics for Build 31, is accessible from the Chromosome Reports, Genome Map Plots, and Summary Statistics links in the "Technical Information about the Assembled Sequence" section below.

We'd like to thank NCBI as well as all the people who collaborated on the data and annotations for this release.

We're pleased to announce the release of a UCSC Genome Browser on the Nov. 2002 rat assembly produced by the Baylor College of Medicine Rat Genome Sequencing Center and the Rat Genome Sequencing Consortium. The sequence was assembled using a hybrid approach that combines the clone by clone and whole genome shotgun methods. A new software program - ATLAS - was developed for this effort. The assembly process resulted in a 6.5-fold coverage of the rat genome, which is estimated to be approximately 2.8 Gigabases in size.

Downloads of the rat data and annotations are available through our ftp site at ftp://genome-archive.cse.ucsc.edu/goldenPath/rnNov2002 or via the archived downloads link. This data contains specific conditions for use. The sequence is also available from the Rat Genome Project website for the Human Genome Sequencing Center at Baylor College of Medicine or from GenBank.

We'd like to thank the Baylor team and the Rat Genome Sequencing Consortium for their collaboration on this project. See the Credits page for a complete list of acknowledgments.. For more information on the rat genome, the assembly process, and the Rat Genome Sequencing Consortium, refer to the website for the Human Genome Sequencing Center at Baylor College of Medicine.

The International Mouse Genome Sequencing Consortium has announced the publication of a high-quality draft sequence of the mouse genome, together with a comparative analysis of the mouse and human genomes. The results from this analysis can be found in the Mouse Genome Browser on this website. The paper appears in the Dec. 5 issue of the journal Nature at http://www.nature.com/nature/mousegenome/. The co-author list includes several members of the UCSC Genome Bioinformatics Group: CBSE Director David Haussler, Research Scientist Jim Kent and research team members Robert Baertsch, Mark Diekhans, Terrence Furey, Angie Hinrichs, Fan Hsu, Donna Karolchik, Krishna Roskin, Matt Schwartz, Charles Sugnet and Ryan Weber.

We've added several new directories of downloadable data to the 28 June 2002 human genome assembly. These directories contain mouse/human alignments of the June 2002 human assembly vs. the Feb. 2002 mouse assembly. You can access these directories from our archived downloads link or or via our ftp site at ftp://genome-archive.cse.ucsc.edu/goldenPath/28jun2002/vsMm2/.

Within the main directory vsMm2 are 3 subdirectories that contain all the alignments (axtAll), alignments filtered to provide only the best alignment for any given region of the human genome (axtBest), and a relatively stringent subset of the axtBest alignments (axtTight). For more information about the format of the alignment files and the methods used to generate the alignments, consult the README.txt file in the vsMm2 directory.

We've rolled out a new version of the Genome Browser - v.16. In addition to several bug fixes, this release contains some interesting new features.

The Table Browser has undergone major enhancements. Users can now restrict their queries by specifying a value or range for any of the fields in a table, and by selecting which fields should be displayed in the output. The Table Browser also provides the ability to do a free-form SQL query on a table and supports several new output formats.

We've extended the capabilities of the DNA retrieval functionality in the Genome Browser and the Table Browser. The new mechanism offers the user several options for configuring the amount and type of sequence region that is retrieved, and options for formatting the sequence output. The retrieval options vary based on the type of table selected.

The Genome Browser's gene prediction tracks now offer a Comparative Sequence link in addition to the predicted protein, mRNA sequence, and genomic sequence links. The Comparative Sequence feature displays annotated codons and translated protein for the region in alignment to another species.

We've recently added a few new tracks/tables to the hg12 and mm2 Browsers. On the latest human assembly (hg12), we now have Chimp Blat and Chimp BAC tracks provided by Ingo Ebersberger, Joshua Bacher, and Svante Pääbo at the Max Planck Institute for Evolutionary Anthropology. Jim Kent at UCSC has generated a new annotation - Gene Boundaries - that shows the boundaries of genes and the direction of transcription as deduced from clustering spliced ESTs and mRNAs against the genome. Daryl Thomas of UCSC has also added SNP tracks to both hg12 and the latest mouse assembly (mm2), based on data from the SNP Consortium and NIH.

We encourage you to experiment with these new features. Comments, questions, and suggestions are always welcome at genome@soe.ucsc.edu.

We have found errors with the RepeatMasker track on the Feb. '02 mouse assembly (mm2). This problem affects the RepeatMasker track and RepeatMasked DNA obtained via the browser's DNA links. It does not affect data downloaded from the browser's downloads page or ftp site. We have replaced the erroneous data set with a corrected version. If you have questions about how this change may affect your project, email genome@soe.ucsc.edu.

Please note that the coordinate range for a portion of the Build 30 chr22 in the UCSC Genome Browser differs from that of NCBI and the Sanger Centre. The latter version contains a 100K bp gap inserted into chr22 just after the centromere. This modification was made after UCSC released the Build 30 assembly in the Genome Browser. All of the chr22 annotations displayed by the UCSC browser are correctly positioned relative to one another. However the coordinates of all features past the centromere will be 100K less than those of NCBI and Sanger. The Ensembl Genome Browser is consistent with the UCSC browser.

We're pleased to announce the release of the Mouse Cons (Human/Mouse Evolutionary Conservation Score) annotation track for the June 2002 human genome assembly. This track allows a user to interactively explore conservation between the human and mouse genomes and identify highly conserved regions. Highest levels of conservation are typically seen over coding exons. High levels of conservation are also frequently associated with noncoding RNA, promoters, other regulatory elements, and pseudogenes.

The Mouse Cons annotation is displayed using a new type of Genome Browser graphical track that plots a continuous function along a chromosome. The conservation levels are calculated over 50bp windows in the human genome that have at least 15 bp aligned to mouse. The score for a window reflects the probability that the level of observed conservation in that 50bp region would occur by chance under neutral evolution. This information is given on a logarithmic scale and displays in the track as "mountain ranges". Details pages associated with the individual peaks in the track provide access to the base level alignments for the whole region and for the individual 50bp windows.

The team that produced this track includes Ryan Weber, Krishna Roskin, Mark Diekhans, Jim Kent, Scott Schwartz, and Webb Miller.

Nature Genetics has just published User's Guide to the Human Genome, a hands-on tutorial for using genome browsers as web tools for browsing and analyzing data from the Human Genome Project and other sequencing efforts. The 3 browsers featured in the tutorial include the UCSC Genome Browser, NCBI's Map Viewer, and the Ensembl Genome Browser. The guide is organized around a collection of step-by-step solutions to 13 typical research questions, and serves as a nice supplement to the documentation materials available on the UCSC Genome Browser web site. The guide is accessible as a link off the Nature Genetics home page at http://www.nature.com.ng.

We've corrected a problem with the Human June 2002 (hg12) cytoBand annotation track that affected chromosome Y. The clones on this chromosome were erroneously pushed from the q-arm onto the p-arm, creating some confusion. The currently available version of the cytoBand data on this website contains this correction.

We've just released several new annotation tracks/tables for the human genome. On the June 2002 assembly, we've added Gene Bounds, UniGene, CpG Islands, Nonhuman mRNA & EST, SNPs, NC160, and GNF Ratio. On the April 2002 assembly we've added a Fgenesh++ Genes track, and have also updated the RepeatMasker track. The Dec. 2001 release now includes a Sanger 22 track.

On the mouse genome, we have 2 new tracks for the Feb. 2002 assembly: TIGR Gene Index and RNA Genes.

We have fixed an error with six of the chr3 contigs in the bigZips/contigAgp.zip file. The following .agp files were corrected: NT_005684.agp, NT_005663.agp, NT_022554.agp, NT_022459.agp, NT_006031.agp, and NT_022419.agp. The chr3.agp file in bigZips/chromAgp.zip was also modified. This change does not affect the .gl files, the .fa files, the lift files, or the annotations. Alignments made on the previous version of chr3 are still good. Updated versions of the contigAgp.zip and chromAgp.zip files were uploaded to our site today. You can download the new versions via ftp at ftp://genome-archive.cse.ucsc.edu/goldenPath/28jun2002 or via the archived downloads link.

The problems with the June 2002 Build 30 (hg12) RepeatMasker track have been resolved. The new RepeatMasker track, along with regenerated Fish Blat and Genscan tracks, are now available in the Browser and through our Downloads link. We've also added a few new annotation tracks for the June 2002 release, and will be adding more over the next 2 weeks.

The latest Genome Browser has 2 new features. We've added filter functionality to the Table Browser, accessible via the Filter Fields button on the Table Browser main page. Also, some of the Dec. 2001 human genome tracks (eg. RefSeq Genes) now have a Comparative Sequence link from the details page that shows annotated codons and translated protein with alignment to the mouse genome.

We've experienced some RepeatMasker problems on Build 30 and are rerunning it. This will directly affect the RepeatMasker track and the masking of the fasta files. The Fish Blat and Genscan tracks may also change slightly once we're redone this. The EST, mRNA, and RefSeq tracks should not be affected. We will also post a new RepeatMasker track for Build 29 (see news item below) as soon as the Build 30 tracks are completed. We apologize for any rework this may cause.

Bulk downloads of the June 2002 Build 30 human genome assembly (hg12) are now available from ftp://genome-archive.cse.ucsc.edu/goldenPath/28jun2002. You can also access the data via the archived downloads link. This initial release of the annotation database download contains a limited set of tables. Additional files will be available for download next week.

The BLAT server and the coordinates conversion feature for human genome assembly Build 30 (hg12) are now functional.

We're pleased to announce the pre-release of a browser for human genome assembly Build 30 from NCBI (UCSC version hg12). This assembly was produced at NCBI based on sequence information submitted into GenBank as of June 28, 2002. Build 30 release notes and statistics will soon be available from the NCBI web site.

Build 30 is an excellent high-quality assembly. It contains nearly 87% finished sequence, and 94%-97% coverage. The sequence coverage of this build is much higher than in previous releases, and there is a high level of correspondence between the sequence and the map. Currently, the human genome project appears to be on track to achieve the goal of finishing at least 95% of the human genome (using Bermuda standards) by April 2003.

UCSC has generated a set of high-level comparisons of the Build 30 draft sequence against various types of information (STS maps, BAC end pairs, and clone overlaps), accessible from the Chromosome Reports and Genome Map Plots links in the "Technical Information about the Assembled Sequence" section below.

A Blat server for Build 30 is not yet available, but should be accessible from this site later this week. Data for the mitochondrial genome and several more annotation tracks will be posted for this release as they become available. Bulk downloads of the hg12 data should be available from this site in a few days.

We've found some problems with the repeat-masking of the Build 29 (hg11) human sequence. We're in the process of replacing the RepeatMasker track, but do not plan to redo the other tracks due to the imminent release of Build 30. Because of this, we advise that you do not use the cross-species tracks for statistical purposes.

The UCSC Genome Bioinformatics home page is sporting an updated interface to accommodate the growing number of organisms supported by the UCSC Genome Browser, BLAT, and Table Browser. The list of assembly versions accessible through each of these tools can now be found on the tool's Gateway page. To reach the Gateway page, choose an organism from the dropdown list on the left sidebar of this page, then click the Browser, BLAT, or Tables link. New organisms will be added to the list in the months ahead.

The UCSC site continues to provide a variety of bulk downloads of a genome assemblies and annotations. The list of downloadable data has been removed from the home page, but is readily available through the Downloads link on the left sidebar. The downloads list can also be accessed directly at http://hgdownload.cse.ucsc.edu/downloads.html or through our ftp site at ftp://hgdownload.cse.ucsc.edu/goldenPath/.

Several new annotation tracks have been added to our site in the past couple weeks. The Feb. 2002 mouse assembly now has tracks for BAC End pairs, Fgenesh++ gene predictions, and AltGenie gene predictions based on Affymetrix's Genie gene-finding software. New to the Apr. 2002 human assembly is the GenMapDB Clones track, which shows placements of BAC clones from the GenMapDB database based on BAC end sequencing information and confirmed using STS markers by Vivian Cheung's lab at U. Penn. We've also changed the Known Genes track name to RefSeq Genes in all assemblies.

This release also includes an updated User's Guide and more detailed documentation on creating & using custom annotation tracks.

Bulk downloads of the April 2002 hg11 human genome assembly (NCBI Build 29) are now available from ftp://genome-archive.cse.ucsc.edu/goldenPath/05apr2002. You can also access the data from the Downloads link in the left sidebar.

We've just released a browser and BLAT server on the latest Build 29 human genome assembly from NCBI (UCSC version hg11). This assembly is based on sequence information submitted into GenBank as of Apr. 5 2002. As with the Dec. 2001 (hg10) release, this assembly was produced at NCBI rather than at UCSC. Consult NCBI's Build 29 release notes and statistics for more information about this release. This assembly contains nearly 75% finished sequence. Currently, the human genome project appears to be on track to achieve the goal of finishing at least 95% of the human genome (using Bermuda standards) by April 2003.

Although the NCBI human genome assembly has been steadily improving over the past year, mapping problems still exist in the current release. Most are small, relatively local rearrangements. Larger scale problems include a rearrangement in the p-arm of Chr16 and several discrepancies in Chr17. Researchers - especially positional cloners - are strongly encouraged to use the tools provided (comparison plots, chromosome reports) to evaluate the accuracy of the assembly in specific regions of interest.

Bulk downloads of the hg11 data should be available from this site in approximately one week. New annotation tracks will be posted as soon as they become available.

Bulk downloads of the February 2002 mouse genome assembly are now available from ftp://genome-archive.cse.ucsc.edu/goldenPath/mmFeb2002. You can also access the data from the archived downloads link.

The February 2002 mouse genome assembly is now available in the browser and for BLAT searching. This assembly was produced at the Whitehead Institute using their Arachne software. We'd like to thank them and the Mouse Genome Sequencing Consortium for providing this assembly, which has specific conditions for use. Bulk downloads of the data should be available in approximately one week. Coordination with mouse genome data access at Ensembl and NCBI is in progress. We'd also like to acknowledge the UCSC team that produced this release: Jim Kent, Terry Furey, Matt Schwartz, Fan Hsu, Yontao Lu, Donna Karolchik, Chuck Sugnet, and Ryan Weber.

Bulk downloads of the November 2001 mouse genome assembly are now available from ftp://genome-archive.cse.ucsc.edu/goldenPath/mmNov2001. You can also access the data from the archived downloads link.

An updated version of the UCSC Genome Browser (v.11) is now available. Along with the v.11 browser, we've released several new annotation tracks on the latest human and mouse assemblies. The new Human Dec. 2001 tracks include: Mouse Synteny, Ensembl, Genscan, CpG Islands, Mouse Blat, Fish Blat, Unigene/SAGE, NCI60 Microarray, GNF Affymetrix Microarray, Rosetta Microarray, and SNPs. An STS Markers track has been added to the Mouse Nov. 2001 browser.

The November 2001 mouse genome assembly is now available for viewing in the browser and for BLAT searching. This assembly was produced at the Sanger Center using the Phusion software developed by Jim Mullikin and Zemin Ning, and was tied to the mouse fingerprint map by Tim Hubbard. We'd like to thank them and the Mouse Genome Sequencing Consortium for providing this assembly, which has specific conditions for use. Bulk downloads of the data will be available in approximately one week. Coordination with mouse genome data access at Ensembl and NCBI is in progress. We'd also like to acknowledge the UCSC team that produced this release: Jim Kent, Terry Furey, Matt Schwartz, Fan Hsu, Yontao Lu, and Donna Karolchik.

A new assembly based on sequence submitted as of Dec. 22 in Genbank (Build 28) is now available in the browser and for BLAT search. This assembly was produced at NCBI rather than UCSC, primarily by Richa Agarwala, Greg Schuler, and Paul Kitts. The NCBI assembly has been steadily improving over the past year. Currently it shows slightly better local order and orientation compared to the UCSC assembly on the same sequence, but somewhat worse tracking of the chromosome level maps. The NCBI assembly has the advantage that it can be generated significantly faster than the UCSC assembly. With the human genome sequencing now in the end game - over two thirds of the human clones are now finished - we feel it more productive to focus worldwide annotation efforts on a single assembly rather than continue producing competing assemblies. We're working with NCBI to improve their map tracking.

Chromosome Reports detailing correspondence with STS map, overlap, and BAC end sequence information are available under the "Technical Information About the Assembled Sequence" section below. This also gives information about the clone map on which the assembled sequence is based.

There are some major enhancements to the browser. The complete user interface settings including track controls, labels, and position are now saved from session to session. You can configure the browser once to your liking and it will stay that way. This feature will only work if cookies are enabled in your browser. If you want to restore the default settings use the reset all button under the main graphic. Also under the main graphic are new controls that move just the start or just the end of the genome window. These are useful for getting exactly the right view without having to do arithmetic on the position. These controls by default will move two guideline units at a time, but you can specify other increments. There's a new page associated with each track. This page is accessible by clicking on the mini-buttons to the left of the track in the main graphic, or by clicking on the new hyperlink associated with the track in the track controls section under the graphic. These pages contain a description of the track and in many cases new controls. The mRNA and EST associated controls let you color or filter the display according to tissue, author, organism, and so forth.

As with any new enhancement there are likely to be a few new bugs too. Many of these have been spotted and fixed already. Please let us know if you find a problem that persists more than a day or two. It's always helpful to include the freeze and genomic position with a problem report.

There is now a link from the known genes details page to the Jackson Lab's MGI Mouse Ortholog when the ortholog is known. Thanks to Carol Bult for her help setting up this link.

A duplications track is now available in the August browser. This track shows duplicate blocks of sequence larger than 1000 bases. The track is hidden by default. To open it look for 'Duplications' in the third row of track controls under the main graphic window, and change the setting to 'dense'. Thanks to Evan Eichler and Jeff Bailey for this track.

Sanger curated gene annotations are now available on chromosome 20. Thanks to Jennifer Ashurst, James Gilbert, and all the annotators at the Sanger Institute.

A new track has been added to the August freeze browser showing haplotype blocks derived from common SNPs on Chromosome 21 by Perlegen, as described in "Common High-Resolution Haplotypes." Patil, N. et. al. Science 294:1719-1723 (2001).

The SNP and Mouse Blat tracks are now available for August. The Mouse Blat track uses a partial assembly of the public whole genome shotgun data courtesy of Whitehead's Arachne program.

The detail web pages for each of the tracks have been updated to reflect the overall look and design of this site. You will now see the familiar blue navigation bar with links to the Browser, BLAT, Downloads, and the FAQ page from each of the track detail pages.

The STS Markers track has been updated on the April and August browsers to now include much more information on the detail page including links to UniSTS and details on the alignments of the markers to the draft sequence. In addition, all known aliases of the markers can be entered in the "position" window, and the corresponding merker will be found and displayed if its location has been determined.

A new FISH Clones track has been added to the April and August browsers. Previously, this information has been included in the STS Markers track. Now, this has been broken out into a separate track with additional information provided on the detail page not previously shown.

The fgenesh++ gene prediction and the cross-species mRNA tracks are now available in the August browser.

The 'DNA' button at the top of the browser has been significantly upgraded. By default it now returns DNA that has repeating elements in lower case and other DNA in upper case. There is also an option to color the DNA output with various tracks. You can have the case and font features such as underline, bold, and italic follow tracks too.

There is now a TIGR Gene Index track in the April 2001 freeze browser. The TIGR Gene Index is based on alignments of assembled ESTs from a number of species. Be sure to click into the track and follow the outside links to the TIGR site, which contains a wealth of information on the genes.

The Acembly track on the August 2001 freeze has been updated to include predictions based on human ESTs and Genbank mRNAs as well as RefSeq human mRNAs. Protein predictions are now also available in the details page for this track. The outside link for this track is also very informative.

You can now share your custom tracks with the community. The easiest way to do this is to construct a link from your own web pages to the browser. Here is an example of a URL for such a link:

   http://genome.ucsc.edu/cgi-bin/hgTracks?
        position=chr22:1-20000&db=hg8&
        hgt.customText=http://genome.ucsc.edu/test.bed

Fresh tab-delimited files from the browser database are now available. They will now be updated automatically every Sunday evening. The table browser queries the database directly, so it is always up to the minute.

Several new sets of gene predictions came in this week. We now have fgenesh++ predictions for the April freeze, and Genscan and Acembly predictions for the August freeze.

Why struggle with massive genomic file downloads when the UCSC Table Browser lets you select exactly the track data desired via a convenient web interface? Major new improvements by Krish Roskin have empowered and simplified this feature available now for the three most recent assemblies.

The October 2000 assembly has been moved to the Archives to make room for the August assembly.

A revised August 2001 freeze assembly is now up. The problems with flipped contigs of finished clones and high levels of sequence duplication are fixed. You can now download this assembly in bulk as well as browse it. Chromosome by chromosome and annotation database files will be following over the next day or two.

You can now convert coordinates between different versions of the draft using a new program, hgCoordConv, by Chuck Sugnet. hgCoordConv attempts to cut out sequences of the original draft and align them to the new draft. When aligning the sequences to the new draft hgCoordConv makes sure that the sequences are in the same order, orientation, and have the correct distances between them.

Chuck has also implemented a SAGE/Unigene track in the browser. This track displays data from the SAGEMap project at NCBI . UniGene cluster sequences are displayed in the browser and colored according to their average SAGE counts over a series of experiments. Selecting one of the UniGene representative sequences displays the SAGE results for UniGene sequences.

Some systematic problems were found in the clone order on the preliminary August 2001 freeze assembly. The sequencing center at Washington University, EBI's Ensembl group, and our group here at UCSC are currently working together to revise the merged BAC clone maps and the assembly process to fix these problems. We hope to update the August browser with a revised assembly soon. Then, after further testing via the browser, we will release the assembled August freeze genome sequence itself.

A preliminary assembly of the August 2001 freeze is now available in the genome browser. Due to significant progress by the mapping and finishing groups of the international public consortium, this assembly is a major improvement over the April 2001 freeze assembly. Imre Vastrik, Ewan Birney and colleagues at Ensembl have computed a merge of BAC clone maps provided by the individual sequencing centers with fingerprint-based maps prepared at Washington University. These merged maps were used for the first time in this August assembly.

The August assembly has successfully passed our internal quality control tests. We will release the sequence and annotations in bulk downloadable form in a week or so, after the external testers have had a chance to further verify it. Meanwhile if you notice any systematic problems please let us know at genome@soe.ucsc.edu. Though the state of the working draft has improved considerably, remember that where you see solid marks in the 'gap' track, the relative order and orientation of flanking contigs is still uncertain. In some cases of complex repeat structure it is also possible that the assembly may be incorrect even in the absence of gaps. Also, sometimes ambiguities in the data cause a BAC clone to be split, with parts of it placed at opposite ends of a run of other clones. Localized errors of this type should be corrected by additional finishing efforts at the individual sequencing centers and should not be reported to UCSC. However, please report any large-scale or systematic problems you detect with this assembly that could have been caused by our data processing.

The tracks available on the August 2001 browser are quite limited at the moment. More tracks will show up over time.

The April 2001 assembly is now the default for the browser. The SNP and Ensembl gene tracks have come in for this version. There is also a new track depicting non-human vertebrate mRNA alignments.

The Sept. 2000 and July 2000 versions of the genome are now only available on our archive site. Please see the link in the blue box to the left for more details.

The August 6 freeze is progressing through the pipeline. We've recently received an updated accession map from Wash U. Ensembl will shortly be integrating this with chromosome specific maps from the sequencing centers. We are still on track for an early September next release.

Meanwhile we've been continuing work on the genome browser. It's now possible to upload your own annotations to be displayed alongside the built-in tracks. Please scroll to the bottom of the browser gateway pages for further information. The browser has also been sped up, particularly on the larger chromosomes by using a 'binning' technique suggested by Lincoln Stein and Richard Durbin.

Tracks continue to be added to the April 2001 browser. Our old friend the Exofish track is back. The blat mouse homology track is now up as well, computed at somewhat more sensitive settings than it was in the December 2000 browser.

We've recently received some significant funding from NHGRI to maintain and extend this site. This has allowed us among other things to hire an artist, Jenny Draper, who is responsible for the new look.

July 13, 2001: fixed bug where some UTRs were mis-annotated in the known genes on the minus strand.

July 13, 2001: fixed bug where some UTRs were mis-annotated in the known genes on the minus strand.

April 5, 2001: chromosome level files (but not contig level files) updated to fix bug where some of the centromeres were misplaced

April 1, 2001: chromosome Y updated to fix a bug that put a large gap between each clone. This bug was limited to the Y chromosome.

April 7, 2001: Affymetrix gene predictions updated and available for bulk download.

January 9, 2001: all files were updated after a bug that had caused some finished clones to be flipped in the assembly was caught and fixed. Our apologies for any inconvenience this has caused.

November 10, 2000: the sequence (.fa) files for two contigs: X/ctg18523/ctg18523.fa and 7/ctg15082/ctg15082.fa were updated. These files had null (zero valued) characters that have been replaced with N characters. These characters were a result of a mismatch between clone sizes in the map and in finished NT contigs. The sequence for chromosomes X and 7, which contain these contigs, has also been updated.

October 11, 2000: chr21.agp and chr22.fa were updated. chr21.agp was a version which went with the UCSC draft assembly rather than the Sanger/NCBI final assembly of this chromosome. chr21.agp and chr21.fa are now in sync. chr22.fa and chr22.agp were also previously out of sync. chr22.fa was obtained from Sanger while chr22.agp had been obtained by NCBI. With this update they are consistent, both NCBI versions. Apologies for any rework this causes you.

October 9, 2000: chr21_random.* and chr22_random.* were removed from the zip-files in the September 5th freeze. These files were relics that should not have been included in the first place. The files chr9_random.agp, chr10_random.agp, chr11_random.agp, chr12_random.agp, chr13_random.agp and chr14_random.agp were updated. There was a bug where the initial field of the initial lines in these files was "(null)" rather than "chrN_random" as it should have been.

September 22, 2000: The zip-files chromFa.zip and chromAgp.zip under the July 17th full data set were updated to fix some duplications of clone contigs that occurred in the chrN_random.agp and chrN_random.fa files contained within these zip-files. These "_random" files are files that contain clone contigs that were mapped to a particular chromosome, but could not be placed at a specific position within that chromosome. They correspond to the "RANDOM" sections of the WashU map. None of the regular chrN.agp or chrN.fa files were affected by this update, nor was any of the information in the contigAgp.zip or contigFa.zip files changed. For convenience, we include two new files, chromRandAgp.zip and chromRandFa.zip, for users who would like to download only the data that has changed. These zips consist of the updated chrN_random.agp and chrN_random.fa files, respectively.

September 4, 2000: The files chromFa.zip and contigFa.zip under the July 17th full data set and the files under July 17th data by individual clone contig were updated to fix some incorrect (null(0)) characters that needed to be replaced by 'n' characters in some Fasta files. The following contig Fasta files on chromosomes 1,8,11,12,16, 17 and 19 were affected:

1/ctg14250/ctg14250.fa
8/ctg17325/ctg17325.fa
8/ctg16307/ctg16307.fa
8/ctg25150/ctg25150.fa
8/ctg15216/ctg15216.fa
11/ctg15548/ctg15548.fa
12/ctg15285/ctg15285.fa
16/ctg16864/ctg16864.fa
16/ctg28077/ctg28077.fa
16/ctg13970/ctg13970.fa
16/ctg25401/ctg25401.fa
16/ctg13336/ctg13336.fa
16/ctg14937/ctg14937.fa
16/ctg16860/ctg16860.fa
16/ctg17794/ctg17794.fa
17/ctg14047/ctg14047.fa
19/ctg16564/ctg16564.fa