From 5ef1e0d2d3d837310ead314f962d2fe37973325c Mon Sep 17 00:00:00 2001
From: Hugo Gillet <hugo.gillet@stud.unibas.ch>
Date: Mon, 5 Dec 2022 22:10:50 +0000
Subject: [PATCH] Update representative.py
---
scripts/representative.py | 66 ++++++++++++++++++++-------------------
1 file changed, 34 insertions(+), 32 deletions(-)
diff --git a/scripts/representative.py b/scripts/representative.py
index fcc7bf8..047a959 100644
--- a/scripts/representative.py
+++ b/scripts/representative.py
@@ -1,15 +1,13 @@
-
+### Made by Hugo Gillet ###
import pandas as pd
-import os
+import os
-'''
+"""
This part of the code take as input a gtf modified file
and return a dictionary of transcripts with best
support level for each gene of the input
-'''
-
-
+"""
def import_gtfSelection_to_df(gtf_modified_file: str) -> pd.DataFrame:
@@ -28,20 +26,24 @@ def import_gtfSelection_to_df(gtf_modified_file: str) -> pd.DataFrame:
"""
pass
if not type(gtf_modified_file) is str:
- raise TypeError("Only str path is allowed")
- df_input = pd.read_csv(gtf_modified_file, sep = '\t', lineterminator = '\n',
-names = ["Gene_mixed", "Transcript", "Support_level", "Na1", "Na2"] )
+ raise TypeError("Only str path is allowed")
+ df_input = pd.read_csv(
+ gtf_modified_file,
+ sep="\t",
+ lineterminator="\n",
+ names=["Gene_mixed", "Transcript", "Support_level", "Na1", "Na2"],
+ )
df_input["Support_level"] = df_input["Support_level"].replace(" ", "")
- df_input["Gene"] = df_input["Gene_mixed"].str.extract('([A-Z]\w{0,})', expand=True)
- df_input["Transcript_number"] = df_input["Gene_mixed"].str.extract('(^\d)', expand=True)
- df_clean = df_input.loc[:, ["Gene", "Transcript","Support_level"]]
- df_clean["Gene"] = df_clean["Gene"].fillna(method = 'ffill')
- df_clean = df_clean.dropna(axis = 0)
+ df_input["Gene"] = df_input["Gene_mixed"].str.extract("([A-Z]\w{0,})", expand=True)
+ df_input["Transcript_number"] = df_input["Gene_mixed"].str.extract(
+ "(^\d)", expand=True
+ )
+ df_clean = df_input.loc[:, ["Gene", "Transcript", "Support_level"]]
+ df_clean["Gene"] = df_clean["Gene"].fillna(method="ffill")
+ df_clean = df_clean.dropna(axis=0)
return df_clean
-
-
def representative_transcripts_inDict(df_gtfSelection: pd.DataFrame) -> pd.DataFrame:
"""Return a dict containing for each gene transcripts
with highest confidence level
@@ -56,22 +58,22 @@ def representative_transcripts_inDict(df_gtfSelection: pd.DataFrame) -> pd.DataF
Raises:
TypeError : Only pandas DataFrame is allowed
"""
- pass
+ pass
if not type(df_gtfSelection) is pd.DataFrame:
raise TypeError("Only pandas DataFrame is allowed")
-
- df_multIndex = df_gtfSelection.set_index(["Gene", "Transcript"])
- #highest support level = 1 , worst = 5, NA = 100
- df_min = df_multIndex[df_multIndex["Support_level"] == df_multIndex["Support_level"].min()]
- df_final = df_min.reset_index(level = "Transcript")
- df_final = df_final.drop(columns = ["Support_level"])
- dict_representative_transcripts = df_final.groupby("Gene")["Transcript"].apply(list).to_dict()
- return dict_representative_transcripts
-
-
-
-def find_repr_by_SupportLevel(intermediate_file: str) -> dict[str,str]:
+ df_min = df_gtfSelection[
+ df_gtfSelection["Support_level"]
+ == df_gtfSelection.groupby("Gene")["Support_level"].transform(min)
+ ]
+ df_final = df_min.drop(columns=["Support_level"])
+ dict_representative_transcripts = (
+ df_final.groupby("Gene")["Transcript"].apply(list).to_dict()
+ )
+ return dict_representative_transcripts
+
+
+def find_repr_by_SupportLevel(intermediate_file: str) -> dict[str, str]:
"""Combine functions import_gtfSelection_to_df()
and representative_transcripts_inDict()
@@ -86,12 +88,12 @@ def find_repr_by_SupportLevel(intermediate_file: str) -> dict[str,str]:
"""
- pass
+ pass
df_gtf = import_gtfSelection_to_df(intermediate_file)
dict_reprTrans = representative_transcripts_inDict(df_gtf)
return dict_reprTrans
-if __name__ == "__main__":
- find_repr_by_SupportLevel()
+if __name__ == "__main__":
+ find_repr_by_SupportLevel()
--
GitLab