def read_score(score_tab, ref):
fa = Fasta(ref)
fa_dict = {}
for i, j in fa.items():
fa_dict[i.split('\t')[0]] = j
score_dict = nested_dict()
with open(score_tab, 'r') as TXT:
for line in TXT:
line = line.strip()
if not line or line.startswith('@'): continue
arr = line.split('\t')
if arr[1] == '-': continue
score_dict[arr[0]][int(arr[2])] = arr[7]
score_dict = score_dict.to_dict()
# print score_dict
reactivity_dict = nested_dict(2, list)
for i, j in score_dict.items():
for p in xrange(1, len(fa_dict[i]) + 1):
if p not in score_dict[i]:
r = 'NULL'
elif score_dict[i][p] == '-1':
r = 'NULL'
else:
r = score_dict[i][p]
reactivity_dict[i]['reactivity_ls'].append(r)
return reactivity_dict.to_dict()
def read_fa(
fa='/Share/home/zhangqf7/gongjing/zebrafish/data/reference/transcriptome/danRer10.refSeq.transcriptome.fa'
):
fa_dict1 = Fasta(fa, key_fn=lambda key: key.split("\t")[0])
fa_dict = {i.split()[0]: j[0:] for i, j in fa_dict1.items()}
print fa_dict.keys()[0:3]
return fa_dict
def calc_nuc_counts(fasta_filename, region_size_min,
region_size_max, verbose):
''' calculate nuc frequencies for normalization.
Returns: dict of nucleotide frequencies.
'''
nuc_counts = defaultdict(Counter)
fasta = Fasta(fasta_filename)
for chrom, seq in fasta.items():
for idx, pos in enumerate(seq):
for region_size in range(region_size_min,
region_size_max + 1):
nucs = seq[idx:idx+region_size]
if len(nucs) < region_size: continue
nuc_counts[region_size][nucs] += 1
return nuc_counts
def read_fa(fa='/Share/home/zhangqf7/gongjing/zebrafish/data/reference/transcriptome/danRer10.refSeq.transcriptome.fa'):
gj.printFuncRun('read_fa')
gj.printFuncArgs()
fa_dict1 = Fasta(fa, key_fn=lambda key:key.split("\t")[0])
fa_dict = {i.split()[0]:j[0:] for i,j in fa_dict1.items()}
print fa_dict.keys()[0:3]
gj.printFuncRun('read_fa')
return fa_dict
def read_fa(
fa='/Share/home/zhangqf5/gongjing/Kethoxal_RNA_structure/data/mm10/transcriptome/mm10_transcriptome.fa'
):
gj.printFuncRun('read_fa')
gj.printFuncArgs()
fa_dict1 = Fasta(fa, key_fn=lambda key: key.split("\t")[0])
fa_dict = {i.split()[0].split('.')[0]: j[0:] for i, j in fa_dict1.items()}
print fa_dict.keys()[0:3]
gj.printFuncRun('read_fa')
return fa_dict
def read_fa(fa=None, species='human', pureID=1):
if fa is None:
if species == 'mouse':
fa = '/home/gongjing/project/shape_imputation/data/ref/mm10/mm10_transcriptome.fa'
if species == 'human':
fa = '/home/gongjing/project/shape_imputation/data/ref/hg38/hg38_transcriptome.fa'
if species == 'human_rRNA':
fa = '/Share2/home/zhangqf5/gongjing/RNA-structure-profile-imputation/data/ref/rRNA_human/ribosomalRNA_4.fa'
if species == 'human_rfam':
fa = '/Share2/home/zhangqf5/gongjing/RNA-structure-profile-imputation/data/ref/Rfam_human/human.dot.dedup.fa'
if species == 'mouse(CIRSseq)':
fa = '/home/gongjing/project/shape_imputation/data/CIRSseq/cirs_txid.fa'
fa_dict1 = Fasta(fa, key_fn=lambda key: key.split("\t")[0])
if pureID:
fa_dict = {
i.split()[0].split('.')[0]: j[0:]
for i, j in fa_dict1.items()
}
else:
fa_dict = {i.split()[0]: j[0:] for i, j in fa_dict1.items()}
print(list(fa_dict.keys())[0:3])
return fa_dict
def search(self, ref_base, pos, alt_base="X"):
var_name = "".join([ref_base, str(pos), alt_base])
fasta_string = self.create_variant_probe_set(var_name=var_name)
with tempfile.NamedTemporaryFile() as fp:
fp.write(fasta_string)
fp.seek(0)
fasta = Fasta(fp.name)
refs = []
alts = []
for k, v in fasta.items():
if "ref" in k:
refs.append(str(v))
else:
alts.append(str(v))
return {"query": var_name, "results": self.genotype_alleles(refs, alts)}
def wig_to_out():
wig = '/home/gongjing/project/shape_imputation/data/CIRSseq/GSE54106_CIRS-seq_Reactivity_combined.wig'
returnval = loadWigHash(wig)
fa = '/home/gongjing/project/shape_imputation/data/CIRSseq/cirs.fa'
fa_dict1 = Fasta(fa)
fa_dict = {}
for i,j in fa_dict1.items():
fa_dict[i.split('|')[0]] = j[0:]
savefn = wig.replace('.wig', '.out')
with open(savefn, 'w') as SAVEFN:
for tx_id in returnval:
if tx_id in fa_dict:
tx_len = len(fa_dict[tx_id][0:])
reactivity_ls = []
for i in range(1, tx_len+1):
if i in returnval[tx_id]:
reactivity_ls.append(returnval[tx_id][i])
else:
reactivity_ls.append('NULL')
SAVEFN.write('\t'.join(map(str, [tx_id, tx_len, '0']+reactivity_ls))+'\n')
def calc_nuc_counts(fasta_filename, region_size_min, region_size_max, verbose):
''' calculate nuc frequencies for normalization.
Returns: dict of nucleotide frequencies.
'''
nuc_counts = defaultdict(Counter)
fasta = Fasta(fasta_filename)
for chrom, seq in fasta.items():
for idx, pos in enumerate(seq):
for region_size in range(region_size_min, region_size_max + 1):
nucs = seq[idx:idx + region_size]
if len(nucs) < region_size: continue
nuc_counts[region_size][nucs] += 1
return nuc_counts
icshape_panpan_1870 = icshape_panpan_1869[1:][0:18] + [
'NULL'
] + icshape_panpan_1869[1:][18:] + ['NULL']
icshape_panpan_1870 = [
np.nan if i == 'NULL' else float(i) for i in icshape_panpan_1870
]
fromPan1870_df = pd.DataFrame({'reactivity': icshape_panpan_1870})
mouse_acces_ls = read_access()
mouse_dot_18S_new, mouse_ct_18S_new = read_mouse_dot()
fa_dict1 = Fasta('../data/rRNA/mouse_ribosomalRNA_4.fa',
key_fn=lambda key: key.split("\t")[0])
fa_dict_rRNA_mouse = {
i.split()[0].split('.')[0]: j[0:]
for i, j in fa_dict1.items()
}
# fa_dict_rRNA_mouse.keys()
df_reactivity_df = pd.DataFrame({
'base': list(fa_dict_rRNA_mouse['18S']),
'dot': list(mouse_dot_18S_new),
'keth-seq': kethoxal_notreat_df['reactivity'],
'icSHAPE': fromPan1870_df['reactivity']
})
df_reactivity_df.head()
### plot diff of single stranded bases
df_reactivity_df = df_reactivity_df[df_reactivity_df['base'] == 'G']
df_reactivity_df = df_reactivity_df[df_reactivity_df['dot'].isin([
'.',