Courbes pour les seuils des premiers consensus et second consensus

src/plot_results.py

+#!/usr/bin/env python3
+
+
 import matplotlib as mpl
 import matplotlib.pyplot as plt
 import numpy as np
@@ -5,14 +8,20 @@ import re
 import sys
 import os
 
+th = [30, 40, 50, 60, 70, 80]
 
 def parse_file(filename, x_scale):
     identity, similarity, equivalence, mismatches = [ [], [] , [], []]
     value = []
+
     with open(filename, 'r') as f:
         cnt= 0
         for line in f:
+            try:
                 i, s, e, m = map(float, line.split(',')[1:])
+            except:
+                i,s,e,m = [0, 0, 0, 0]
+                print("Missing values replaced by 0, 0, 0, 0")
             identity.append(i)
             similarity.append(s)
             equivalence.append(e)
@@ -21,26 +30,53 @@ def parse_file(filename, x_scale):
     return identity, similarity, equivalence, mismatches
 
 
-th = [30, 40, 50, 60, 70, 80]
+def plot_from_folder(args, target, labels, folder):
+
     ls, li, le, lm = [], [], [], []
-for tool_file in os.scandir(sys.argv[1]):
-    if ".txt" in tool_file.name:
-        i, s, e, m = map(np.array, parse_file(tool_file, th))
+    for tool_file in args:
+        if folder != "":
+            cond =  ".txt" in tool_file.name
+            filename = folder+tool_file.name
+        else:
+            cond = ".txt" in tool_file
+            filename= tool_file
+        if cond :
+            i, s, e, m = map(np.array, parse_file(filename, th))
             li.append(i)
             ls.append(s)
             le.append(e)
             lm.append(m)
-
-
+    fig = plt.figure(figsize=(20,10))
     ll = [ls, li, le, lm]
-cnt = 0
+    graph_cnt = 0
     target_att = ["identity", "similarity", "equivalence", "mismatches"]
     for graph in ll:
         plt.clf()
+        tab_cnt=0
         for tab in graph:
             if (len(tab) > 1):
-            plt.plot(th, tab, 'bo', label=cnt)
+                plt.plot(th, tab, 'o', label=labels[tab_cnt])
+                
+                #    plt.plot(th,tab, label=labels[tab_cnt])
             else:
-            plt.plot(th,[tab for x in range(len(th))])
-    plt.savefig(sys.argv[2]+"_" + target_att[cnt] + ".png")
-    cnt+=1
+                plt.plot(th,[tab for x in range(len(th))], label=labels[graph_cnt])
+            tab_cnt+=1
+        legend = plt.legend(shadow=True, fontsize='x-large')
+        plt.savefig(target+"_" + target_att[graph_cnt] + ".png", format="png",)
+        graph_cnt+=1
+
+if __name__ == "__main__":
+    
+    if (len(sys.argv) < 3):
+        exit(-1)
+    if (len(sys.argv) == 3):
+        print(sys.argv)
+        folder_it = os.scandir(sys.argv[2])
+        name= [ x.name for x in folder_it]
+        #folder_it= os.scandir(sys.argv[2])
+        plot_from_folder(folder_it, sys.argv[1], name, folder=sys.argv[2])
+        exit(0)
+    lb= []
+    for elem in th:
+         lb.append("th "+str(elem))
+    plot_from_folder(sys.argv[2:] ,sys.argv[1], sys.argv[2:], folder="")

src/workflow/Snakefile

@@ -5,6 +5,7 @@ THRESHOLD = ["30", "40", "50", "60", "70", "80"]
 REGION = ["100", "200", "500", "1000", "2000"]
 DEPTH = ["10", "20", "50", "100", "150"]
 ALIGN = ["muscle", "mafft"]
+GRAPH_TYPES = ["identity", "similarity", "equivalence", "mismatches"]
 
 STAT_HEADER= "threshold, percent_identity, percent_similarity, total_equivalence, total_mismatches\n "
 
@@ -77,8 +78,6 @@ rule exonerate_ref_result:
 #          "echo {STAT_HEADER} >> {output} \n for FILE in {input.files}; do exonerate --bestn 1 -Q dna -E -m a:g --showalignment false --showsugar false --showvulgar false --showcigar false  --ryo {resources.format} --verbose 0 -q {input.ref} -t $FILE >> {output}  2>/dev/null; done" #
 
 
-
-
 rule plottable_exonerate_meta_t2:
      input: files=expand(rules.final_consensus.output, align="muscle", value=THRESHOLD, allow_missing=True) , ref=expand("{data}/seq_selectes_region/region_seq_r{region}.fasta", data=DATA, allow_missing=True)
      output: "consensus/plot/{region}/{depth}/{threshold}/aln_stats_meta.txt"
@@ -93,8 +92,22 @@ rule plottable_exonerate_tools_t2:
      conda: "envs/exonerate.yaml"
      shell: "./exonerate_stats.sh {output} {input.ref} {input.consensus_per_tool}"
 
-rule plot_t2:
-     input: expand(rules.plottable_exonerate_tools_t2.output, region=REGION, depth=DEPTH, threshold=THRESHOLD, tool=TOOL),
-     	     expand(rules.plottable_exonerate_meta_t2.output, region=REGION, depth=DEPTH, threshold=THRESHOLD, tool=TOOL)
+
+rule plot_t1:
+     input: 
+     	     expand(rules.plottable_exonerate_meta_t2.output, threshold=THRESHOLD, allow_missing=True),
+     params: target_folder="consensus/plot/{region}/{depth}"
+     output: expand("consensus/plot/{region}/{depth}/graph_r{region}_d{depth}_{graph}.png", graph=GRAPH_TYPES, allow_missing=True)
+     shell: "./plot_results.py {params.target_folder}/graph_r{wildcards.region}_d{wildcards.depth} {input}"
      
 
+rule plot_t2:
+     input: expand(rules.plottable_exonerate_tools_t2.output, tool=TOOL, allow_missing=True),
+     	     rules.plottable_exonerate_meta_t2.output
+     params: target_folder="consensus/plot/{region}/{depth}/{threshold}/"
+     output: expand("consensus/plot/{region}/{depth}/{threshold}/graph_r{region}_d{depth}_t1_{threshold}_{graph}.png", graph=GRAPH_TYPES, allow_missing=True)
+     shell: "./plot_results.py {params.target_folder}graph_r{wildcards.region}_d{wildcards.depth}_t1_{wildcards.threshold}  {input}"
+
+rule make_plots:
+     input: expand(rules.plot_t2.output, region=REGION, depth=DEPTH, threshold=THRESHOLD),
+     	    expand(rules.plot_t1.output, region=REGION, depth=DEPTH)
\ No newline at end of file