import pysam
import numpy as np
from matplotlib import pyplot as plt
from matplotlib.backends.backend_pdf import PdfPages
import glob
import multiprocessing as mp
def insertSize(bam):
sam = pysam.AlignmentFile(bam)
insize_list = []
for read in sam.fetch():
insize = np.abs(read.template_length) # 插入片段大小
insize_list.append(insize)
return insize_list
def drawHist(insize_list,ids):
insize_list = [i for i in insize_list if i <=500]
fig, ax1 = plt.subplots(figsize=(20, 15))
plt.title(ids)
ax1.hist(insize_list, bins=len(set(insize_list))) # bins 直方图柱子数量
ax1.set_xlabel("insert size", size=18)
ax1.set_ylabel("Fre", size=18)
my_x_ticks = list(range(0, 500, 20))
plt.xticks(my_x_ticks)
plt.show()
plt.close()
insize_list = insertSize(bam)
drawHist(insize_list, sampleid)
Tip: cfDNA中来自肿瘤细胞的ctDNA 片段比正常细胞的cfDNA片段要短, ctDNA集中在 134-144bp之间,并且呈现10bp的周期性;而来自正常细胞的cfDNA片段集中在165bp 。
使用 picard画插图片段
/usr/bin/java -Xmx10g -jar path/to/Software/Picard-Tools-2.1.0/picard.jar CollectInsertSizeMetrics I= yourBam.bam O=sample.insertsize.txt H=sample.insertsize.pdf