par.add_argument('--o', required = True, help = 'name of output file')
par.add_argument('--uplimit', type = float, help = 'top n sequences to calculate')
args = par.parse_args()
database_path = '/home/anthill/fzheng/home/searchDB/support_bc-30-sc-correct-20141022/others'
odir = os.getcwd()
ldir = General.createLocalSpace()
outfh = open(ldir + '/' + args.o, 'w')
uplimit = args.uplimit
nseq = 0
for match_line in open(args.m):
if (uplimit != None) and (nseq == uplimit):
break
match_line = match_line.strip()
indices = Analyze.index_from_match(match_line)
index1, index2 = indices[args.n[0]], indices[args.n[1]]
target_pds = match_line.split()[1]
targetid = General.getBase( General.removePath(match_line.split()[1]) )
env_dict = database_path + '/' + targetid[1:3] + '/' + targetid + '.freedom.db'
db = shelve.open(env_dict, 'r')
# extract post-processed pdb files from target_pds
resfile = database_path + '/' + targetid[1:3] + '/' + General.changeExt( General.removePath(target_pds), 'post.res')
allres = open(resfile).read().splitlines()
resid1, resid2 = allres[index1], allres[index2]
resid1, resid2 = resid1[0] + ',' + resid1[1:], resid2[0] + ',' + resid2[1:]
fields = ['sumcond', 'crwdnes', 'freedom', 'phi', 'psi', 'aa']
outfh.write(targetid + '\t')
if not resid1 in db:
if args.conres != None:
tempfile2 = General.changeExt(args.m, 'seqcontext.fasta2')
tempfh2 = open(ldir + '/' + tempfile2, 'w')
# output file names
nr_matchf = args.outh + '_' + args.m
nr_seqf = General.changeExt(nr_matchf, 'seq')
nr_env = None
oenv = General.changeExt(args.m, args.env)
if os.path.isfile(oenv):
nr_env = General.changeExt(nr_matchf, args.env)
# write a custom .fasta file
for match in matches:
match_region_indices = Analyze.index_from_match(match)
central_index = match_region_indices[args.cres - 1]
match_id = General.getBase( General.removePath( match.split()[1] ) )
fullsequence = database[match_id]
if central_index - args.wd < 1:
seqcontext = fullsequence[0:(2 * args.wd + 1)]
elif central_index + args.wd > len(fullsequence):
seqcontext = fullsequence[-(2 * args.wd + 1):]
else:
seqcontext = fullsequence[(central_index - args.wd - 1):(central_index + args.wd)]
tempfh.write('>match:'+str(matchind)+'\n'+seqcontext+'\n')
if args.conres != None:
con_index = match_region_indices[args.conres -1]
if con_index - args.wd < 1:
