Glossary of major routines and variables used in the RNAGA ---------------------------------------------------------- Major routines: pop1st - Call "gtstr" to obtain initial set of structures for each sequence. maturation - Apply GA iteration to the initial population until every strucure in the population reach to the desired stability. gtcnfrm - Generate 2n new structures by performing crossover and mutation operations on current populationof n structures. Mutate every structure in the population to obtain n new structures. Select n pair of structures based on the structure conservation score. Do crossover for each pair to obtain another n new structures. hmlgy - Compute conservation score for each structure. A structure is represented as a set of stems. Each stem assign a weight. A stem get full credit if there is an equivalent stem (see manuscript) in another sequence, otherwise, no credit to the stem. The conservation score of a structure is the sum of its stem scores. hmlgyy - Compute the adjusted conservation score for each structure. The computation is similar to hmlgy except a stem may get partial credit of weight. gtwt - Compute the weight for each stem in a structure. Called by hmlgy and hmlgyy. nodcmpr - Determine whether a stem has an equivalent stem in another sequence, and also compute the score for the stem. Called by hmlgy and hmlgyy. fndcmstr - Find potential common structures that satisfy certain conditions in structural conservation score and stability. gtpopH - Select next generation based on structural conservation scores and also the diversity of structures. ritcmstr - Write out the predicted common structures. The structure is represented as a set of stems (region table). mutat - Perform mutation operation. The stem with low conservation score has better chance to be selected to mutate. xovr - Perform uniform crossover operation. The structure with high conservation has better chance to be a parental structure. Global variables: mxnseq - maximum number of sequences in sequence file lmax - maximum length of sequence mxnp - maximum number of stuctures in a population mxnp2 - maximum number of structures generated during GA iteration which is 3*mxnp mxnr - maximum number of regions (stems) in a structure Hcrtval - minimum conservation score for a structure to be considered as a potential common structure Hcrtval2 - minimum conservation score for a structure to be a common structure mseq(mxnseq,lmax) - original multiple sequences returned from gtmseq lenseq(mxnseq) - array for the length of sequences seq(lmax) - hold single sequence 1st structure information ireg(mxnseq,mxnp2,mxnr) - array for the 5' of stem of structures generated during GA iteration jreg(mxnseq,mxnp2,mxnr) - array for the 3' of stem of structures generated during GA iteration rsz(mxnseq,mxnp2,mxnr) - array for the length of stem of structures generated during GA iteration struc(mxnseq,mxnp2,2) - array for the number of stems and the energy of structures generated during GA iteration irHm(mxnseq,mxnp5,mxnr) - array for the 5' of stem of predicted common structures jrHm(mxnseq,mxnp5,mxnr) - array for the 3' of stem of predicted common structures rszHm(mxnseq,mxnp5,mxnr) - array for the length of stem of predicted common structures strHm(mxnseq,mxnp5,2) - array for the number of stems and the energy of predicted common structures nstrHm(mxnseq) - array for the number of predicted common structures regscr(mxnseq,mxnp5,mxnr) - array for the stem conservation score Hscore(mxnseq,mxnp2) - array for the structural conservation score in which a stem get either full credir of weight or none Hmscr(mxnseq,mxnp) - array for the structural conservation score in which a stem may get partialcredir of weight