fixed trinity bug

Author: Tobias Hofmann

recipe/meta.yaml

@@ -1,4 +1,4 @@
-{% set version = "1.1.9" %}
+{% set version = "1.1.10" %}
 
 package:
   name: secapr
@@ -7,7 +7,7 @@ package:
 source:
   fn: secapr_{{ version }}.tar.gz
   url: https://github.com/AntonelliLab/seqcap_processor/archive/v{{ version }}.tar.gz
-  sha256: ca80b1d8fa99e572ffe0a7a7c09f77cfa0f2701cb4ffeb21a6b1a4468640b059
+  sha256: 175830fa23619a374719c12340504fb50f9b00d6136b49c2a925873ecda687d1
 
 build:
   skip: True  # [not py27]

secapr.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 1.1
 Name: secapr
-Version: 1.1.0+45.ge31aab9.dirty
+Version: 1.1.0+47.gd364be5.dirty
 Summary: Process sequence-capture fastq files into alignments for phylogenetic analyses
 Home-page: https://github.com/AntonelliLab/seqcap_processor
 Author: Tobias Hofmann

secapr/assemble_reads.py

@@ -131,7 +131,7 @@ def main(args):
                     print ('#' * 50)
                     print ("Processing sample %s" %sample_id)
                     if assembler == "trinity":
-                        assembly_trinity(forward,backward,sample_output_folder,sample_id,cores,min_length)
+                        assembly_trinity(forward,backward,sample_output_folder,sample_id,cores,min_length,max_memory)
                         contig_count_df = get_trinity_stats(sample_output_folder,sample_id,sample_contig_count_dict)
                         cleanup_trinity_assembly_folder(sample_output_folder,sample_id)
                         print ("#" * 50)

secapr/locus_selection.py

@@ -248,8 +248,10 @@ def extract_best_loci(subfolder_file_dict,sample_bam_dict,output_folder,n,thresh
         convert_to_bam = 'samtools view -Sb %s > %s' %(sam_output_file,bam_output_file)
         os.system(convert_to_bam)
         sorted_bam_out = os.path.join(output_subfolder_dict[sample],"%s_%s_selected_loci_sorted.bam" %(sample,input_type))
-        sort_bam = 'samtools sort %s > %s' %(bam_output_file,sorted_bam_out)
+        sort_bam = 'samtools sort %s %s' %(bam_output_file,sorted_bam_out.replace('.bam',''))
         os.system(sort_bam)
+        index_bam = 'samtools index %s' %(sorted_bam_out)
+        os.system(index_bam)
     return target_loci
 
 

src/find_good_loci_for_each_sample_from_readcov_info.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Created on Wed Jun 27 10:47:27 2018
+
+@author: tobias
+"""
+
+import numpy as np
+import pandas as pd
+from functools import reduce
+
+read_cov_overview = '/Users/tobias/GitHub/seqcap_processor/data/processed/remapped_reads/average_cov_per_locus.txt'
+read_cov_overview_df = pd.read_csv(read_cov_overview,sep='\t')
+
+loci_names = read_cov_overview_df['locus'].values
+good_exons = []
+sample_exon_count_dict = {}
+for sample in read_cov_overview_df.columns:
+    if not sample =='locus':
+        values = read_cov_overview_df[sample].values
+        num_loci_high_coverage = len(values[values>3])
+        sample_exon_count_dict.setdefault(sample,num_loci_high_coverage)
+        good_loci_names = loci_names[values>3]
+        good_exons.append(good_loci_names)
+
+loci_present_in_all_samples = reduce(np.intersect1d, (good_exons))
+
+
+exons_per_sample = list(sample_exon_count_dict.values())
+np.mean(exons_per_sample)
+np.std(exons_per_sample)