Upload parsing py script for clustering

scripts/extractdiffanalysis.py

+#!/usr/bin/env python3
+"""
+Author: Joran Schoorlemmer
+Student nr: 1004586
+Description: Write log2fold and padj values from tsv files into two csv files
+Usage: python3 extractdiffanalysis.py <listofsamplefiles.txt> \
+<log2foldoutputfilename.txt> <padjoutputfilename.txt>
+"""
+
+
+# Import
+from sys import argv
+
+
+# Functions
+def extractlogfold(csv_file):
+    """Extract fold changes from csv file
+
+    csv_file: str, name of input sample csv file
+
+    fold_dict: dict, dict with gene names as keys and l2f changes as values
+    """
+    with open(csv_file, 'r') as logf:
+        logf.readline()
+        fold_dict = {} # init dict
+        for line in logf:
+            line.strip()
+            line_info = line.split('\t') # write tsv line to list
+            gene_name = line_info[0]
+            log2fold = float(line_info[2])
+            fold_dict[gene_name] = log2fold
+        return fold_dict
+
+
+def extractpadj(csv_file):
+    """Extract padj values from csv file
+
+    csv_file: str, name of input sample csv file
+
+    p_dict: dict, dict with gene names as keys and l2f changes as values
+    """
+    with open(csv_file, 'r') as logf:
+        logf.readline()
+        p_dict = {} # init dict
+        for line in logf:
+            line.strip()
+            line_info = line.split('\t') # write tsv line to list
+            gene_name = line_info[0]
+            padj = float(line_info[6].strip())
+            p_dict[gene_name] = padj
+        return p_dict
+
+
+def writecsv(sample_dict, out_file):
+    """Write sample dict into csv file
+
+    sample_dict: dict, dict with samples as keys and dicts with gene names as 
+    keys as values
+    out_file: str, filename of output csv file
+    """
+    with open(out_file, 'w') as out_f:
+        out_f.write('Gene\t')
+        gene_names = set()
+        # write samples and find gene names
+        for sample in sample_dict:
+            gene_names.update(list(sample_dict[sample].keys()))
+            out_f.write(str(sample[0:4])+"\t") # parse sample name
+        out_f.write('\n')
+        for gene in gene_names:
+            out_f.write(str(gene)+"\t")
+            for sample in sample_dict:
+                try: 
+                    out_f.write(str(sample_dict[sample][gene])+'\t')
+                except: # if gene is not present in csv, write NA
+                    out_f.write('NA\t')
+            out_f.write('\n')
+
+
+# Main
+def main(csv_names, out_file_fold, out_file_p):
+    """Run functions and loop trough sample files
+
+    csv_names: str, filename of txt file with sample file names
+    out_file_fold: str, filename of l2f output csv file
+    out_file_p: str, filename of padj output csv file
+    """
+    with open(csv_names,'r') as f_names:
+        # init dicts
+        dict_all_fold = {}
+        dict_all_padj = {}
+        # loop trough files
+        for fn in f_names:
+            fn = fn.strip('\n')
+            fold_dict = extractlogfold(fn)
+            p_dict = extractpadj(fn)
+            dict_all_fold[fn] = fold_dict
+            dict_all_padj[fn] = p_dict
+        # write csv files
+        writecsv(dict_all_fold, out_file_fold)
+        writecsv(dict_all_padj, out_file_p)
+
+
+# Run script
+if __name__ == "__main__":
+    main(argv[1], argv[2], argv[3])
\ No newline at end of file