def set_groupdata_dict(self, groupdata_file: str):
self.groupdata_file = groupdata_file
self.groupdata_digest_name = Utilities.filename_only(self.groupdata_file).replace(".groupdata", "")
groupdata_df = pd.read_table(self.groupdata_file, sep="\t", header="infer", names=["sample_name", "group_name"])
self.groupdata_dict = {i: sorted(set(
groupdata_df.loc[groupdata_df["group_name"] == i, ["sample_name"]])) for i in sorted(
set(groupdata_df["group_name"]))}
self.raw_all_sample_names_list = sorted(set(groupdata_df["sample_name"]))
def split(self, output_dir: str):
output_dir = Utilities.ends_with_slash(output_dir)
os.makedirs(output_dir, exist_ok=True)
# Note: the dataframe must have only index and value columns
for sample_col_name in list(self.pivot_df):
sample_name = Utilities.filename_only(sample_col_name).split(
"_")[0]
sample_file_name = "{}{}.tsv".format(output_dir, sample_name)
self.pivot_df[sample_col_name].reset_index().rename(
columns={
sample_col_name: self.value_col_name
}).to_csv(sample_file_name, sep="\t", header=True, index=False)
self._sample_names_list.append(sample_file_name)
def evaluate_sampledata():
import os
import subprocess
import pandas as pd
from meta.scripts.Utilities import Utilities
import re
#
df = pd.DataFrame(columns=["sample_name", "sample_path"])
for dir_mask in [
"/data2/bio/ecoli_komfi/raw_reads/*",
"/data2/bio/ecoli_komfi/raw_reads2/*"
]:
data_1 = [
i.strip() for i in subprocess.getoutput(
"ls -d {}R1*.fastq* | sort".format(dir_mask)).split("\n")
]
data_12 = [
"{a}\t{b}".format(a=i, b=i.replace("R1", "R2"))
if os.path.isfile(i.replace("R1", "R2")) else ""
for i in data_1
]
sample_names_list = [
re.sub("_S.*$", "", Utilities.filename_only(i)) for i in data_1
]
df = pd.concat([
df,
pd.DataFrame.from_dict({
"sample_name": sample_names_list,
"sample_path": data_12
})
],
axis=0,
ignore_index=True)
#
df["group_id"] = "group_id"
#
os.makedirs(ProjectDescriber.directory, exist_ok=True)
df.loc[:,
["sample_name", "group_id"]].to_csv(ProjectDescriber.groupdata,
sep='\t',
index=False,
header=False)
df.loc[:, ["sample_name", "sample_path"]].to_csv(
ProjectDescriber.sampledata, sep='\t', index=False, header=False)
subprocess.getoutput(
"sed -i 's|\"||g' {}".format(ProjectDescriber.sampledata)
) # Tab-containing columns items are flanked by '"'
