clean repo

deconvolution/main/to_d2_graph.py

@@ -69,6 +69,7 @@ def main():
     dprint("constructing d2 graph from barcode graph")
     d2g.construct_from_barcodes(neighbor_threshold=args.edge_divergence_threshold, clique_mode=clique_mode, threads=args.threads, verbose=args.verbose)
     dprint("[debug] d2 graph constructed")
+    dprint("[debug] %d nodes %d edges" % (len(d2g.nodes()), len(d2g.edges())))
     
     # d2g.save(f"{args.output_prefix}.tsv")
     nx.write_gexf(d2g, f"{args.output_prefix}.gexf")

interval_graphs_acc_sens.sh

@@ -10,7 +10,7 @@ do
             echo "n=$n d=$d m=$m m_dev=$dev"
 
             rm  -f snake_exec/simu_0_bar_n"$n"_d"$d"_m"$m"-dev0.gexf && 
-            snakemake -s Snakefile_data_simu  --config n=$n d=$d m=$m m_dev=$dev >/dev/null 2>&1
+            snakemake --cores 1 -s Snakefile_data_simu  --config n=$n d=$d m=$m m_dev=$dev >/dev/null 2>&1
             cd snake_exec && source CMD_acc_sens && cd .. 
 
 done

medvedev_repo_scripts/evaluate_barcode_graph.py

 # compares a barcode graph to ground truth
-# input: [tested.gpickle] [ground_truth.gpickle]
+import sys
 
+if len(sys.argv) < 3:
+    exit("input: [tested.gpickle] [ground_truth.gpickle]")
 
-import sys
 from collections import Counter
 from networkx import nx
 
+
 print("loading tested graph")
 tested = nx.read_gpickle(sys.argv[1])
 

medvedev_repo_scripts/ground_truth_barcode_graph.py

@@ -5,7 +5,7 @@
 # input: [reads_file.fastq.gz] [output_prefix]
 #
 # example:
-# zcat ~/data/10x-ecoli/fastqs/ecoli.fq.gz | python ground_truth_barcode_graph.py /dev/stdin graph
+# zcat ~/data/10x-ecoli/fastqs/ecoli.fq.gz | python ground_truth_barcode_graph.py /dev/stdin graph [min_overlap_length] [max_nb_overlaps_per_mol] [equi]
 
 import sys, gzip
 from Bio.SeqIO.QualityIO import FastqGeneralIterator
@@ -16,6 +16,20 @@ from collections import Counter
 
 reads_file = sys.argv[1]
 overlap_size = 9000
+if len(sys.argv) > 3:
+    overlap_size = int(sys.argv[3])
+    print("using custom overlap size %d" % overlap_size)
+
+max_overlaps = None # limits the number of overlaps we record per molecule (keeping only the longest ones)
+if len(sys.argv) > 4:
+    max_overlaps = int(sys.argv[4])
+    print("using custom maximal number of overlaps per molecule %d" % max_overlaps)
+
+equi = False # among the neighbors of a molecule, force to have as many molecules before than as after  (i.e. prevents a molecule from only overlapping with those after itself) - useful in conjunction with max_overlaps
+if len(sys.argv) > 5:
+    equi = True
+    print("asking for equi before/after overlap repartition")
+
 
 if reads_file.endswith('.gz'):
     opener = gzip.open
@@ -48,8 +62,8 @@ for title, seq, qual in FastqGeneralIterator(opener(reads_file,"rt")):
 
     #print(read)
     # WARNING: here if it crashes, just uncomment one of the two lines and comment the other
-    #chrom, posA, posB = read.split('_')[1:4] #specific to reference names with a "_"
-    chrom, posA, posB = read.split('_')[0:3] #specific to reference names without "_"
+    chrom, posA, posB = read.split('_')[1:4] #specific to reference names with a "_"
+    #chrom, posA, posB = read.split('_')[0:3] #specific to reference names without "_"
     posA, posB = map(int,[posA,posB])
     #print(posA,posB,barcode)
 
@@ -154,6 +168,7 @@ for barcode in barcodes_labels:
     long_label = ';'.join(barcodes_labels[barcode])
     bG.add_node(barcode, label=long_label)
 
+overlaps = defaultdict(list)
 for i, molecule in enumerate(molecule_extents):
     chrom, posA, posB = molecule_extents[molecule]
     results = tree.find_overlap(posA,posB)
@@ -161,18 +176,43 @@ for i, molecule in enumerate(molecule_extents):
     for otherposA, otherposB, other_molecule in results:
         other_molecule = vals[other_molecule]
         otherchrom, otherposA_2, otherposB_2 = molecule_extents[other_molecule]
-        assert(otherposA == otherposA_2 and otherposB == otherposB_2)
-        if chrom == otherchrom and overlap(posA,posB,otherposA,otherposB) > overlap_size:
-            #print(posA,posB,molecule,"-",otherposA, otherposB, other_molecule)
-            mG.add_edge(molecule,other_molecule)
-            barcode      =       molecule.split("-")[0]
-            otherbarcode = other_molecule.split("-")[0]
+        assert(otherposA == otherposA_2 and otherposB == otherposB_2) # verifies the interval tree result
+        overlap_length = overlap(posA,posB,otherposA,otherposB)
+        if chrom == otherchrom and overlap_length >= overlap_size:
             edge_label= '%s:%d-%d %s:%d-%d' %(chrom,posA,posB,otherchrom,otherposA,otherposB)
-            bG.add_edge(barcode,otherbarcode, label=edge_label)
+            overlaps[molecule]       += [(overlap_length, other_molecule, edge_label)]
     if i % 1000 == 1:
         sys.stderr.write(".")
         sys.stderr.flush()
 
+for molecule in overlaps:
+    nb_needed_overlaps = float('inf') if max_overlaps is None else max_overlaps #- mG.degree(molecule)
+    nb_before, nb_after = 0,0 # some stats over the neighbors of a molecule
+    _, posA, posB = molecule_extents[molecule]
+    for (overlap_length, other_molecule, edge_label) in sorted(overlaps[molecule])[::-1]:
+        
+        # examine whether that candidate neighbor starts before or after that current molecule
+        _, otherposA, otherposB = molecule_extents[other_molecule]
+        if otherposA > posA:
+            if equi and nb_after >= max_overlaps//2: continue
+            nb_after += 1
+        else:
+            if equi and nb_before >= max_overlaps//2: continue
+            nb_before += 1
+
+        #print(posA,posB,molecule,"-",otherposA, otherposB, other_molecule)
+        mG.add_edge(molecule,other_molecule)
+        barcode      =       molecule.split("-")[0]
+        otherbarcode = other_molecule.split("-")[0]
+        bG.add_edge(barcode,otherbarcode, label=edge_label)
+
+        nb_needed_overlaps -= 1
+        if nb_needed_overlaps == 0:
+            break
+
+    if nb_before != max_overlaps//2 or nb_after != max_overlaps//2:
+        print(molecule,"before/after %d/%d" % (nb_before,nb_after))
+
 print("writing graphs")
 print(len(bG.nodes()),"nodes",len(bG.edges()),"edges for the barcode graph")
 print(len(mG.nodes()),"nodes",len(mG.edges()),"edges for the molecule graph")