From 69515d326906f3a815c419e16ebeb96ca961c676 Mon Sep 17 00:00:00 2001
From: Hugo Madge Leon <hugo.madgeleon@stud.unibas.ch>
Date: Wed, 7 Dec 2022 08:35:19 +0000
Subject: [PATCH] Hope this doesn't break
---
frag_package/fragmentation.py | 64 -------------------
frag_package/setup.py => setup.py | 0
.../main.py | 0
.../utils.py | 0
4 files changed, 64 deletions(-)
delete mode 100644 frag_package/fragmentation.py
rename frag_package/setup.py => setup.py (100%)
rename {frag_package => terminal-fragment-selector}/main.py (100%)
rename {frag_package => terminal-fragment-selector}/utils.py (100%)
diff --git a/frag_package/fragmentation.py b/frag_package/fragmentation.py
deleted file mode 100644
index 80762e2..0000000
--- a/frag_package/fragmentation.py
+++ /dev/null
@@ -1,64 +0,0 @@
-import re
-
-import numpy as np
-import pandas as pd
-
-
-def fasta_process(fasta_file):
- with open(fasta_file, "r") as f:
- lines = f.readlines()
-
- ident_pattern = re.compile('>(\S+)')
- seq_pattern = re.compile('^(\S+)$')
-
- genes = {}
- for line in lines:
- if ident_pattern.search(line):
- seq_id = (ident_pattern.search(line)).group(1)
- elif seq_id in genes.keys():
- genes[seq_id] += (seq_pattern.search(line)).group(1)
- else:
- genes[seq_id] = (seq_pattern.search(line)).group(1)
- return genes
-
-def fragmentation(fasta_file, counts_file, mean_length, std, a_prob, t_prob, g_prob, c_prob):
- fasta = fasta_process(fasta_file)
- seq_counts = pd.read_csv(counts_file, names = ["seqID", "count"])
-
- # nucs = ['A','T','G','C']
- # mononuc_freqs = [0.22, 0.25, 0.23, 0.30]
- nuc_probs = {'A':a_prob, 'T':t_prob, 'G':g_prob, 'C':c_prob} # calculated using https://www.nature.com/articles/srep04532#MOESM1
-
- term_frags = []
- for seq_id, seq in fasta.items():
- counts = seq_counts[seq_counts["seqID"] == seq_id]["count"]
- for _ in range(counts):
- n_cuts = int(len(seq)/mean_length)
-
- # non-uniformly random DNA fragmentation implementation based on https://www.nature.com/articles/srep04532#Sec1
- # assume fragmentation by sonication for NGS workflow
- cuts = []
- cut_nucs = np.random.choice(list(nuc_probs.keys()), n_cuts, p=list(nuc_probs.values()))
- for nuc in cut_nucs:
- nuc_pos = [x.start() for x in re.finditer(nuc, seq)]
- pos = np.random.choice(nuc_pos)
- while pos in cuts:
- pos = np.random.choice(nuc_pos)
- cuts.append(pos)
-
- cuts.sort()
- cuts.insert(0,0)
- term_frag = ""
- for i, val in enumerate(cuts):
- if i == len(cuts)-1:
- fragment = seq[val+1:cuts[-1]]
- else:
- fragment = seq[val:cuts[i+1]]
- if mean_length-std <= len(fragment) <= mean_length+std:
- term_frag = fragment
- if term_frag == "":
- continue
- else:
- term_frags.append(term_frag)
- return term_frags
-
diff --git a/frag_package/setup.py b/setup.py
similarity index 100%
rename from frag_package/setup.py
rename to setup.py
diff --git a/frag_package/main.py b/terminal-fragment-selector/main.py
similarity index 100%
rename from frag_package/main.py
rename to terminal-fragment-selector/main.py
diff --git a/frag_package/utils.py b/terminal-fragment-selector/utils.py
similarity index 100%
rename from frag_package/utils.py
rename to terminal-fragment-selector/utils.py
--
GitLab