From 1144c1874ef369cdf2b4e6b01d3eb4f2eb6f7c1e Mon Sep 17 00:00:00 2001
From: Fontrodona Nicolas <nicolas.fontrodona@ens-lyon.fr>
Date: Fri, 18 Dec 2020 12:45:19 +0100
Subject: [PATCH] src/bed_handler/bed_resize.py: allow to resize the bed file
---
src/bed_handler/bed_resize.py | 102 ++++++++++++++++++++++++++++++++++
1 file changed, 102 insertions(+)
create mode 100644 src/bed_handler/bed_resize.py
diff --git a/src/bed_handler/bed_resize.py b/src/bed_handler/bed_resize.py
new file mode 100644
index 0000000..8c10bda
--- /dev/null
+++ b/src/bed_handler/bed_resize.py
@@ -0,0 +1,102 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description: The goal of this script is to resize a bed
+"""
+
+import pandas as pd
+import lazyparser as lp
+from doctest import testmod
+from .config import OutputBed
+
+
+def resize_row_from_start(rowo: pd.Series, size: int, resize_from: str
+ ) -> pd.Series:
+ """
+ Resize the bed feature from start inside a bed row.
+
+ :param rowo: A bed row
+ :param size: The maximum size the feature must have
+ :param resize_from: The coordinate for which we want to resize (default \
+ "start")
+ :return: The row resized
+
+ >>> crow = pd.Series({"#ref": 10, "start": 10, "end": 20, "id": 1,
+ ... "score": 1, "strand": "+"})
+ >>> resize_row_from_start(crow, 5, "start").to_dict()
+ {'#ref': 10, 'start': 10, 'end': 15, 'id': 1, 'score': 1, 'strand': '+'}
+ >>> resize_row_from_start(crow, 5, "end").to_dict()
+ {'#ref': 10, 'start': 15, 'end': 20, 'id': 1, 'score': 1, 'strand': '+'}
+ >>> resize_row_from_start(crow, 20, "start").to_dict()
+ {'#ref': 10, 'start': 10, 'end': 20, 'id': 1, 'score': 1, 'strand': '+'}
+ >>> crow = pd.Series({"#ref": 10, "start": 50, "end": 60, "id": 1,
+ ... "score": 1, "strand": "-"})
+ >>> resize_row_from_start(crow, 5, "start").to_dict()
+ {'#ref': 10, 'start': 55, 'end': 60, 'id': 1, 'score': 1, 'strand': '-'}
+ >>> resize_row_from_start(crow, 5, "end").to_dict()
+ {'#ref': 10, 'start': 50, 'end': 55, 'id': 1, 'score': 1, 'strand': '-'}
+ """
+ row = rowo.copy()
+ row_strand = row["strand"]
+ d= {"+": "-", "-": "+"}
+ row_strand = d[row_strand] if resize_from == "end" else row_strand
+ if row["end"] - row["start"] <= size:
+ return row
+ if row_strand == "+":
+ row['end'] = row['start'] + size
+ else:
+ row['start'] = row['end'] - size
+ return row
+
+
+def update_bed(df_bed: pd.DataFrame, size: int,
+ resize_from: str) -> pd.DataFrame:
+ """
+ Resize each feature in a bed dataframe.
+
+ :param df_bed: The dataframe corresponding to a bed to resize
+ :param size: The maximum size the feature must have
+ :param resize_from: The coordinate for which we want to resize (default \
+ "start")
+ :return: The dataframe resized
+
+ >>> cdf = pd.DataFrame({"#ref": [1, 1], "start": [10, 50],
+ ... "end": [20, 60], "id": [1, 2], "strand": ["+", "-"]})
+ >>> update_bed(cdf, 5, "start")
+ #ref start end id strand
+ 0 1 10 15 1 +
+ 1 1 55 60 2 -
+ >>> update_bed(cdf, 5, "end")
+ #ref start end id strand
+ 0 1 15 20 1 +
+ 1 1 50 55 2 -
+ """
+ list_s = [
+ resize_row_from_start(df_bed.iloc[i, :], size, resize_from)
+ for i in range(df_bed.shape[0])
+ ]
+ return pd.DataFrame(list_s)
+
+
+@lp.parse(bed="file", size="size > 0", resize_from=["start", "end"])
+def bed_resizer(bed: str, size: int, outfile: str,
+ resize_from: str = "start") -> None:
+ """
+ Resize bed features inside a bed file from their start or stop \
+ position (according to their strand).
+
+ :param bed: A bed file with the features to resize
+ :param size: The maximum size the feature must have
+ :param outfile: The output bed name
+ :param resize_from: The coordinate for which we want to resize (default \
+ "start")
+ """
+ df = pd.read_csv(bed, sep="\t")
+ ndf = update_bed(df, size, resize_from)
+ ndf.to_csv(OutputBed.output / outfile, sep="\t", index=False)
+
+
+if __name__ == "__main__":
+ bed_resizer()
\ No newline at end of file
--
GitLab