Add plotting of deamination frequencies

Author: merszym

conf/process.config

@@ -176,6 +176,17 @@ process {
        ext.args = "${params.doublestranded ? 'doublestranded' : ''}"
     }
 
+    withName:"PLOT_DEAM" {
+        publishDir = [
+            [
+                path: "${outdir}/stats",
+                mode: "copy",
+                saveAs: { "${meta.RG}_04_deamination_positions/${meta.Family}.${meta.Species}.plot.jpg" },
+                pattern: "*.jpg",
+            ],
+        ]
+    }
+
     withName:"SAMTOOLS_MPILEUP" {
         ext.args = "--output-BP-5 --no-output-ends --no-output-ins --no-output-del --no-output-ins --no-output-del --min-BQ 0  -R --ff 0"
         publishDir = [

modules/local/pandas_plot_deam/main.nf

@@ -0,0 +1,16 @@
+process PLOT_DEAM{
+    container (workflow.containerEngine ? "merszym/pandas_plt:nextflow" : null)
+    tag "${meta.id}"
+    label 'local'
+
+    input:
+    tuple val(meta), path(tsv), path(confidence)
+
+    output:
+    path("*.jpg") , emit: jpg
+
+    script:
+    """
+    plot_deam_patterns.py ${tsv} ${meta.Family}.${meta.Species}
+    """
+}

modules/local/pandas_plot_deam/resources/usr/bin/plot_deam_patterns.py

@@ -0,0 +1,43 @@
+#! /usr/bin/env python3
+
+import pandas as pd
+import seaborn as sns
+import matplotlib.pyplot as plt
+import sys
+
+df = pd.read_csv(sys.argv[1], sep='\t')
+title = sys.argv[2]
+
+df = df.melt(id_vars=['Sub'])
+
+bases = list(set(df['Sub']))
+palette = {x:'lightgrey' for x in bases}
+palette.update({'C->T':'red', "G->A":'black'})
+
+#now plot the rate
+# fix the positions first
+
+df['Position in Sequence'] = df.apply(lambda x: int(x.variable) if int(x.variable) > 0 else int(x.variable)+21, axis=1)
+df['Frequency'] = df['value'].apply(lambda x: float(x))
+df['Substitution'] = df['Sub']
+
+g = sns.lineplot(
+    data = df, 
+    x='Position in Sequence',
+    y='Frequency', 
+    hue='Substitution',
+    palette=palette
+)
+
+ticks = list(range(0,11))
+ticks.extend(list(range(-10,0)))
+
+# Update the x-ticks
+g.set_ylabel("Frequency [%]")
+g.set_xticks(list(range(0,21)))
+g.set_xticklabels(ticks)
+g.axvline(x=10.5, ymax=0.5, ls="--", color='grey')
+g.set_title(title)
+
+plt.tight_layout()
+plt.savefig(f"{title}_plot.jpg", dpi=300)

workflows/08_deamination_stats.nf

@@ -2,6 +2,7 @@ include { BAM_DEAM_STATS as BAM_DEAM_BEST  } from '../modules/local/bam_deam_sta
 include { BAM_DEAM_STATS as BAM_DEAM_FIXED } from '../modules/local/bam_deam_stats'
 include { MASK_DEAMINATION                 } from '../modules/local/mask_deamination'
 include { SAMTOOLS_MPILEUP                 } from '../modules/local/samtools_mpileup'
+include { PLOT_DEAM                        } from '../modules/local/pandas_plot_deam'
 
 workflow deamination_stats {
     take: best
@@ -22,6 +23,8 @@ workflow deamination_stats {
         }
         .set{ best }
 
+        ch_positions = BAM_DEAM_BEST.out.positions
+
         // Now for the fixed
 
         BAM_DEAM_FIXED( fixed )
@@ -40,10 +43,17 @@ workflow deamination_stats {
         }
         .set{ fixed }
 
+        ch_positions = ch_positions.mix(BAM_DEAM_FIXED.out.positions)
+        
+        // Plot deamination rates
+        PLOT_DEAM(ch_positions)
+
+        // And continue with the fixed files...
+
         MASK_DEAMINATION( fixed )
 
         //
-        // And create Mpileups
+        // Create Mpileups
         //
 
         SAMTOOLS_MPILEUP( MASK_DEAMINATION.out.bam )