From c978b20705fb8165232505c6b57ebe247a404584 Mon Sep 17 00:00:00 2001
From: nicolas-zimmermann <nicolaszimmermann.snv@gmail.com>
Date: Thu, 24 Oct 2019 17:09:25 +0200
Subject: [PATCH] added select_best_path function

---
 debruijn/debruijn.py | 50 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 39 insertions(+), 11 deletions(-)

diff --git a/debruijn/debruijn.py b/debruijn/debruijn.py
index 8981acd..88fe456 100644
--- a/debruijn/debruijn.py
+++ b/debruijn/debruijn.py
@@ -8,6 +8,8 @@ from networkx import algorithms
 
 def read_fastq(fichier):
     """
+    Returns an iterator object that retrieves only the nucleic sequences of a
+    fastq file.
     Arguments:
         fichier, str: path to fastq file
 
@@ -22,6 +24,7 @@ def read_fastq(fichier):
 
 def cut_kmer(seq, k):
     """
+    Returns an iterator that returns k-mers of k-size of a sequence
     Arguments:
         seq, str: a sequence
         k, int: k-mer size, must be shorter than len(seq)
@@ -42,10 +45,10 @@ def build_kmer_dict(fichier, k):
         hash_table, dict: dictionnary with key = k-mer as str
                           and value count of k-mer occurence
     """
-    hash_table = {}
+    hash_table = {} # initialise empty hash table
     it_fastq = read_fastq(fichier)
-    for seq in it_fastq:
-        it_kmer = cut_kmer(seq, k)
+    for seq in it_fastq: # for each sequence
+        it_kmer = cut_kmer(seq, k) # count each occurence of k-mer
         for kmer in it_kmer:
             try:
                 hash_table[kmer]
@@ -58,6 +61,8 @@ def build_kmer_dict(fichier, k):
 
 def build_graph(hash_table):
     """
+    Returns a graph from a hash table
+
     Arguments:
         hash_table, dict: dictionnary obtained with build_kmer_dict() function
     Return:
@@ -71,6 +76,8 @@ def build_graph(hash_table):
 
 def get_starting_nodes(graph):
     """
+    Returns the list of starting nodes of a graph
+
     Arguments:
         graph, nx.DiGraph: de Bruijn tree
 
@@ -79,24 +86,27 @@ def get_starting_nodes(graph):
     """
     starting_nodes = []
     for node in graph:
-        if graph.in_degree(node) == 0:
+        if graph.in_degree(node) == 0: # if count of input edge == 0
             starting_nodes.append(node)
 
     return starting_nodes
 
 def std(values):
     """
+    Computes standard deviation from a list of value
     Arguments:
         values, list: list of values
 
     Returns :
         standard deviation of the 'values' data list
     """
-    return stdev(float(values))
+    return statistics.stdev(float(values))
+
 
 
 def get_sink_nodes(graph):
     """
+
     Arguments:
         graph, nx.DiGraph: de Bruijn tree
 
@@ -105,7 +115,7 @@ def get_sink_nodes(graph):
     """
     sink_nodes = []
     for node in graph:
-        if graph.out_degree(node) == 0:
+        if graph.out_degree(node) == 0: # if count of output edge == 0
             sink_nodes.append(node)
 
     return sink_nodes
@@ -154,8 +164,27 @@ def remove_paths(graph, paths, delete_entry_node=False, delete_sink_node=False):
     return graph
 
 
-def select_best_path():
-    pass
+def select_best_path(graph, paths, path_len, mean_weights,
+                     delete_entry_node=False, delete_sink_node=False):
+    """
+    
+    """
+    max_weight = max(mean_weights)
+    heaviest = [i for i, j in enumerate(mean_weights) if j == mean_weights]
+    if len(heaviest) > 1:
+        max_len = max(path_lengths)
+        longest = [i for i in heaviest if path_len[i] == max_len]
+        if len(longest) > 1:
+            Random.seed(9001)
+            best = random.choice[longest]
+        else:
+            best = longest[0]
+    else:
+        best = heaviest[0]
+    paths.pop(best)
+
+    return remove_paths(graph, paths, delete_entry_node, delete_sink_node)
+
 
 def fill(text, width=80):
     """Split text with a line return to respect fasta format"""
@@ -173,7 +202,6 @@ def save_contigs(tuples, outname):
             i += 1
             outfile.write(">contig_{} len={}\n".format(i, duo[1]))
             outfile.write("{}\n".format(fill(duo[0])))
-
     return
 
 
@@ -192,9 +220,9 @@ def get_contigs(graph, starting_nodes, sink_nodes):
         for sink_node in sink_nodes:
             if algorithms.has_path(graph, starting_node, sink_node) == True:
                 path = algorithms.shortest_path(graph, starting_node, sink_node)
-                contig = path[0]
+                contig = path[0] # base of the contig is seq of the first node
                 for i in range(len(path)-1):
-                    contig += path[i+1][-1]
+                    contig += path[i+1][-1] # adds last char of node
                 contigs.append((contig, len(contig)))
     
     return contigs