def get_df(self, window=100):
print("building GC content")
data = tools._base_content(self.filename, window, "GC")
names = self.fasta.names
lengths = self.fasta.lengths
GC = [np.nanmean(data[name]) for name in names]
nreads = [0] * len(GC)
covStats = [0] * len(GC)
if self.mode == "canu":
for i, comment in enumerate(self.fasta.comments):
read = [x for x in comment.split() if x.startswith("reads")][0]
covStat = [
x for x in comment.split() if x.startswith("covStat")
][0]
read = read.split("=")[1]
covStat = covStat.split("=")[1]
nreads[i] = int(read)
covStats[i] = float(covStat)
#if self.bamfile
df = pd.DataFrame({
"GC": list(GC),
"length": lengths,
"name": names,
"nread": nreads,
"covStat": covStats
})
# deal with the bamfile
if self.bam:
bam_df = self.bam.get_df()
bam_df = bam_df.query("flag in [0,16]")
bam_df.set_index("qname", inplace=True)
chrom_name = bam_df.loc[self.fasta.names]["rname"]
df["chromosome"] = list(chrom_name)
self._df = df.copy()
return df
def get_gc(self, window=100):
data = tools._base_content(self.filename, window, "GC")
names = self.fasta.names
lengths = self.fasta.lengths
GC = [100 * np.nanmean(data[name]) for name in names]
return GC