From 132757a573da495050d144b83aae67fd3afbce7a Mon Sep 17 00:00:00 2001
From: LauraU123 <laura.urbanska@stud.unibas.ch>
Date: Fri, 11 Nov 2022 09:39:32 +0100
Subject: [PATCH] updated poisson sampling
---
scripts/poisson_sampling.py | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/scripts/poisson_sampling.py b/scripts/poisson_sampling.py
index 92537e8..60d043d 100644
--- a/scripts/poisson_sampling.py
+++ b/scripts/poisson_sampling.py
@@ -16,13 +16,14 @@ output: csv file with gene id and count
'''
-def transcript_sampling(total_transcript_number, tsv_file, output_csv):
- df = pd.read_csv(tsv_file, sep='\t', lineterminator='\n', names=["id", "level"]))
+def transcript_sampling(total_transcript_number, csv_file, output_csv):
+ df = pd.read_csv(csv_file, sep='\t', lineterminator='\n', names=["id", "level"])
levels = []
-
+ sums = df['level'].tolist()
+ total = sum(sums)
+ normalized = total_transcript_number/total
for expression_level in df['level']:
-
- poisson_sampled = np.random.poisson(total_transcript_number*expression_level)
+ poisson_sampled = np.random.poisson(expression_level*normalized)
levels.append(poisson_sampled)
transcript_numbers = pd.DataFrame({'id': df['id'],'count': levels})
--
GitLab