chenhaonan
/
quartet_dna_quality_control_wes_big_pipeline
forked from renluyao/quartet_dna_quality_control_wes_big_pipeline
Watch 1
Star 0
Fork 0
Code Issues 0 Pull Requests 0 Releases 1 Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
16 Commits
1 Branch
Tree: ddead72efc
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'ddead72efc'
${ noResults }
quartet_dna_quality_control.../codescripts/replicates_consensus.py

228 lines
12KB
Raw Blame History 

  1. from __future__ import division

  2. from glob import glob

  3. import sys, argparse, os

  4. import fileinput

  5. import re

  6. import pandas as pd

  7. from operator import itemgetter

  8. from collections import Counter

  9. from itertools import islice

  10. from numpy import *

  11. import statistics

  12. 

  13. # input arguments

  14. parser = argparse.ArgumentParser(description="this script is to merge mendelian and vcfinfo, and extract high_confidence_calls")

  15. 

  16. parser.add_argument('-prefix', '--prefix', type=str, help='prefix of output file',  required=True)

  17. parser.add_argument('-vcf', '--vcf', type=str, help='merged multiple sample vcf',  required=True)

  18. 

  19. 

  20. args = parser.parse_args()

  21. prefix = args.prefix

  22. vcf = args.vcf

  23. 

  24. # input files

  25. vcf_dat = pd.read_table(vcf)

  26. 

  27. # all info

  28. all_file_name = prefix + "_all_summary.txt"

  29. all_sample_outfile = open(all_file_name,'w')

  30. all_info_col = 'CHROM\tPOS\tREF\tALT\tLCL5_consensus_calls\tLCL5_detect_number\tLCL5_same_diff\tLCL6_consensus_calls\tLCL6_detect_number\tLCl6_same_diff\tLCL7_consensus_calls\tLCL7_detect_number\tLCL7_same_diff\tLCL8_consensus_calls\tLCL8_detect_number\tLCL8_same_diff\n'

  31. all_sample_outfile.write(all_info_col)

  32. 

  33. # filtered info

  34. vcf_header = '''##fileformat=VCFv4.2

  35. ##fileDate=20200501

  36. ##source=high_confidence_calls_intergration(choppy app)

  37. ##reference=GRCh38.d1.vd1

  38. ##INFO=<ID=VOTED,Number=1,Type=Integer,Description="Number mendelian consisitent votes">

  39. ##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">

  40. ##contig=<ID=chr1,length=248956422>

  41. ##contig=<ID=chr2,length=242193529>

  42. ##contig=<ID=chr3,length=198295559>

  43. ##contig=<ID=chr4,length=190214555>

  44. ##contig=<ID=chr5,length=181538259>

  45. ##contig=<ID=chr6,length=170805979>

  46. ##contig=<ID=chr7,length=159345973>

  47. ##contig=<ID=chr8,length=145138636>

  48. ##contig=<ID=chr9,length=138394717>

  49. ##contig=<ID=chr10,length=133797422>

  50. ##contig=<ID=chr11,length=135086622>

  51. ##contig=<ID=chr12,length=133275309>

  52. ##contig=<ID=chr13,length=114364328>

  53. ##contig=<ID=chr14,length=107043718>

  54. ##contig=<ID=chr15,length=101991189>

  55. ##contig=<ID=chr16,length=90338345>

  56. ##contig=<ID=chr17,length=83257441>

  57. ##contig=<ID=chr18,length=80373285>

  58. ##contig=<ID=chr19,length=58617616>

  59. ##contig=<ID=chr20,length=64444167>

  60. ##contig=<ID=chr21,length=46709983>

  61. ##contig=<ID=chr22,length=50818468>

  62. ##contig=<ID=chrX,length=156040895>

  63. '''

  64. consensus_file_name = prefix + "_consensus.vcf"

  65. consensus_outfile = open(consensus_file_name,'w')

  66. consensus_col = '#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER\tINFO\tFORMAT\tLCL5_consensus_call\tLCL6_consensus_call\tLCL7_consensus_call\tLCL8_consensus_call\n'

  67. consensus_outfile.write(vcf_header)

  68. consensus_outfile.write(consensus_col)

  69. 

  70. # function

  71. def decide_by_rep(vcf_list):

  72. 	consensus_rep = ''

  73. 	gt = [x.split(':')[0] for x in vcf_list]

  74. 	gt_num_dict = Counter(gt)

  75. 	highest_gt = gt_num_dict.most_common(1)

  76. 	candidate_gt = highest_gt[0][0]

  77. 	freq_gt = highest_gt[0][1]

  78. 	if freq_gt >= 2:

  79. 		consensus_rep = candidate_gt

  80. 	else:

  81. 		consensus_rep = 'inconGT'

  82. 	return consensus_rep

  83. 

  84. def consensus_call(vcf_info_list):

  85. 	consensus_call = '.'

  86. 	detect_number = '.'

  87. 	same_diff = '.'

  88. 	# pcr

  89. 	SEQ2000 = decide_by_rep(vcf_info_list[0:3])

  90. 	XTen_ARD = decide_by_rep(vcf_info_list[18:21])

  91. 	XTen_NVG = decide_by_rep(vcf_info_list[21:24])

  92. 	XTen_WUX = decide_by_rep(vcf_info_list[24:27])

  93. 	Nova_WUX = decide_by_rep(vcf_info_list[15:18])

  94. 	pcr_sequence_site = [SEQ2000,XTen_ARD,XTen_NVG,XTen_WUX,Nova_WUX]

  95. 	pcr_sequence_dict = Counter(pcr_sequence_site)

  96. 	pcr_highest_sequence = pcr_sequence_dict.most_common(1)

  97. 	pcr_candidate_sequence = pcr_highest_sequence[0][0]

  98. 	pcr_freq_sequence = pcr_highest_sequence[0][1]

  99. 	if pcr_freq_sequence > 3:

  100. 		pcr_consensus = pcr_candidate_sequence

  101. 	else:

  102. 		pcr_consensus = 'inconSequenceSite'

  103. 	# pcr-free

  104. 	T7_WGE = decide_by_rep(vcf_info_list[3:6])

  105. 	Nova_ARD_1 = decide_by_rep(vcf_info_list[6:9])

  106. 	Nova_ARD_2 = decide_by_rep(vcf_info_list[9:12])

  107. 	Nova_BRG = decide_by_rep(vcf_info_list[12:15])

  108. 	sequence_site = [T7_WGE,Nova_ARD_1,Nova_ARD_2,Nova_BRG]

  109. 	sequence_dict = Counter(sequence_site)

  110. 	highest_sequence = sequence_dict.most_common(1)

  111. 	candidate_sequence = highest_sequence[0][0]

  112. 	freq_sequence = highest_sequence[0][1]

  113. 	if freq_sequence > 2:

  114. 		pcr_free_consensus = candidate_sequence

  115. 	else:

  116. 		pcr_free_consensus = 'inconSequenceSite'

  117. 	gt = [x.split(':')[0] for x in vcf_info_list]

  118. 	gt = [x.replace('./.','.') for x in gt]

  119. 	detected_num = 27 - gt.count('.')

  120. 	gt_remain = [e for e in gt if e not in {'.'}]

  121. 	gt_set = set(gt_remain)

  122. 	if len(gt_set) == 1:

  123. 		same_diff = 'same'

  124. 	else:

  125. 		same_diff = 'diff'

  126. 	tag = ['inconGT','inconSequenceSite']

  127. 	if (pcr_consensus == pcr_free_consensus) and (pcr_consensus not in tag):

  128. 		consensus_call = pcr_consensus

  129. 	elif (pcr_consensus in tag) and (pcr_free_consensus in tag):

  130. 		consensus_call = 'notAgree'				

  131. 

  132. 	else:

  133. 		consensus_call = 'notConsensus'

  134. 	return consensus_call, detected_num, same_diff

  135. 

  136. 	elif (pcr_consensus in tag) and (pcr_free_consensus in tag):

  137. 		consensus_call = 'filtered'				

  138. 	elif ((pcr_consensus == './.') or (pcr_consensus in tag)) and ((pcr_free_consensus not in tag) and (pcr_free_consensus != './.')):

  139. 		consensus_call = 'pcr-free-speicifc'				

  140. 	elif ((pcr_consensus != './.') or (pcr_consensus not in tag)) and ((pcr_free_consensus in tag) and (pcr_free_consensus == './.')):

  141. 		consensus_call = 'pcr-speicifc'			

  142. 	elif (pcr_consensus == '0/0') and (pcr_free_consensus == '0/0'):

  143. 		consensus_call = '0/0'								

  144. 	else:

  145. 		consensus_call = 'filtered'

  146. 

  147. 

  148. for row in vcf_dat.itertuples():

  149. # length

  150. #alt

  151. 	if ',' in row.ALT:

  152. 		alt = row.ALT.split(',')

  153. 		alt_len = [len(i) for i in alt]

  154. 		alt_max = max(alt_len)

  155. 	else:

  156. 		alt_max = len(row.ALT)

  157. #ref

  158. 	ref_len = len(row.REF)

  159. 	if (alt_max > 50) or (ref_len > 50):

  160. 		pass

  161. 	else:

  162. # consensus

  163. 		lcl5_list = [row.Quartet_DNA_BGI_SEQ2000_BGI_LCL5_1_20180518,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL5_2_20180530,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL5_3_20180530, \

  164. 						row.Quartet_DNA_BGI_T7_WGE_LCL5_1_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL5_2_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL5_3_20191105, \

  165. 						row.Quartet_DNA_ILM_Nova_ARD_LCL5_1_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL5_2_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL5_3_20181108, \

  166. 						row.Quartet_DNA_ILM_Nova_ARD_LCL5_4_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL5_5_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL5_6_20190111, \

  167. 						row.Quartet_DNA_ILM_Nova_BRG_LCL5_1_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL5_2_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL5_3_20180930, \

  168. 						row.Quartet_DNA_ILM_Nova_WUX_LCL5_1_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL5_2_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL5_3_20190917, \

  169. 						row.Quartet_DNA_ILM_XTen_ARD_LCL5_1_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL5_2_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL5_3_20170403, \

  170. 						row.Quartet_DNA_ILM_XTen_NVG_LCL5_1_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL5_2_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL5_3_20170329, \

  171. 						row.Quartet_DNA_ILM_XTen_WUX_LCL5_1_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL5_2_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL5_3_20170216]

  172. 		lcl6_list = [row.Quartet_DNA_BGI_SEQ2000_BGI_LCL6_1_20180518,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL6_2_20180530,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL6_3_20180530, \

  173. 						row.Quartet_DNA_BGI_T7_WGE_LCL6_1_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL6_2_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL6_3_20191105, \

  174. 						row.Quartet_DNA_ILM_Nova_ARD_LCL6_1_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL6_2_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL6_3_20181108, \

  175. 						row.Quartet_DNA_ILM_Nova_ARD_LCL6_4_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL6_5_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL6_6_20190111, \

  176. 						row.Quartet_DNA_ILM_Nova_BRG_LCL6_1_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL6_2_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL6_3_20180930, \

  177. 						row.Quartet_DNA_ILM_Nova_WUX_LCL6_1_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL6_2_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL6_3_20190917, \

  178. 						row.Quartet_DNA_ILM_XTen_ARD_LCL6_1_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL6_2_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL6_3_20170403, \

  179. 						row.Quartet_DNA_ILM_XTen_NVG_LCL6_1_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL6_2_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL6_3_20170329, \

  180. 						row.Quartet_DNA_ILM_XTen_WUX_LCL6_1_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL6_2_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL6_3_20170216]

  181. 		lcl7_list = [row.Quartet_DNA_BGI_SEQ2000_BGI_LCL7_1_20180518,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL7_2_20180530,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL7_3_20180530, \

  182. 						row.Quartet_DNA_BGI_T7_WGE_LCL7_1_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL7_2_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL7_3_20191105, \

  183. 						row.Quartet_DNA_ILM_Nova_ARD_LCL7_1_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL7_2_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL7_3_20181108, \

  184. 						row.Quartet_DNA_ILM_Nova_ARD_LCL7_4_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL7_5_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL7_6_20190111, \

  185. 						row.Quartet_DNA_ILM_Nova_BRG_LCL7_1_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL7_2_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL7_3_20180930, \

  186. 						row.Quartet_DNA_ILM_Nova_WUX_LCL7_1_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL7_2_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL7_3_20190917, \

  187. 						row.Quartet_DNA_ILM_XTen_ARD_LCL7_1_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL7_2_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL7_3_20170403, \

  188. 						row.Quartet_DNA_ILM_XTen_NVG_LCL7_1_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL7_2_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL7_3_20170329, \

  189. 						row.Quartet_DNA_ILM_XTen_WUX_LCL7_1_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL7_2_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL7_3_20170216]

  190. 		lcl8_list = [row.Quartet_DNA_BGI_SEQ2000_BGI_LCL8_1_20180518,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL8_2_20180530,row.Quartet_DNA_BGI_SEQ2000_BGI_LCL8_3_20180530, \

  191. 						row.Quartet_DNA_BGI_T7_WGE_LCL8_1_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL8_2_20191105,row.Quartet_DNA_BGI_T7_WGE_LCL8_3_20191105, \

  192. 						row.Quartet_DNA_ILM_Nova_ARD_LCL8_1_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL8_2_20181108,row.Quartet_DNA_ILM_Nova_ARD_LCL8_3_20181108, \

  193. 						row.Quartet_DNA_ILM_Nova_ARD_LCL8_4_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL8_5_20190111,row.Quartet_DNA_ILM_Nova_ARD_LCL8_6_20190111, \

  194. 						row.Quartet_DNA_ILM_Nova_BRG_LCL8_1_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL8_2_20180930,row.Quartet_DNA_ILM_Nova_BRG_LCL8_3_20180930, \

  195. 						row.Quartet_DNA_ILM_Nova_WUX_LCL8_1_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL8_2_20190917,row.Quartet_DNA_ILM_Nova_WUX_LCL8_3_20190917, \

  196. 						row.Quartet_DNA_ILM_XTen_ARD_LCL8_1_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL8_2_20170403,row.Quartet_DNA_ILM_XTen_ARD_LCL8_3_20170403, \

  197. 						row.Quartet_DNA_ILM_XTen_NVG_LCL8_1_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL8_2_20170329,row.Quartet_DNA_ILM_XTen_NVG_LCL8_3_20170329, \

  198. 						row.Quartet_DNA_ILM_XTen_WUX_LCL8_1_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL8_2_20170216,row.Quartet_DNA_ILM_XTen_WUX_LCL8_3_20170216]

  199. 		# LCL5

  200. 		LCL5_consensus_call, LCL5_detected_num, LCL5_same_diff = consensus_call(lcl5_list)

  201. 		# LCL6

  202. 		LCL6_consensus_call, LCL6_detected_num, LCL6_same_diff = consensus_call(lcl6_list)

  203. 		# LCL7

  204. 		LCL7_consensus_call, LCL7_detected_num, LCL7_same_diff = consensus_call(lcl7_list)

  205. 		# LCL8

  206. 		LCL8_consensus_call, LCL8_detected_num, LCL8_same_diff = consensus_call(lcl8_list)

  207. 		# all data

  208. 		all_output = row._1 + '\t' + str(row.POS) + '\t' + row.REF + '\t' + row.ALT + '\t' +  LCL5_consensus_call + '\t' + str(LCL5_detected_num) + '\t' + LCL5_same_diff + '\t' +\

  209. 						 LCL6_consensus_call + '\t' + str(LCL6_detected_num) + '\t' + LCL6_same_diff + '\t' +\

  210. 						 LCL7_consensus_call + '\t' + str(LCL7_detected_num) + '\t' + LCL7_same_diff + '\t' +\

  211. 						 LCL8_consensus_call + '\t' + str(LCL8_detected_num) + '\t' + LCL8_same_diff + '\n'

  212. 		all_sample_outfile.write(all_output)

  213. 		#consensus vcf

  214. 		one_position = [LCL5_consensus_call,LCL6_consensus_call,LCL7_consensus_call,LCL8_consensus_call]

  215. 		if ('notConsensus' in one_position) or (((len(set(one_position)) == 1) and ('./.' in set(one_position))) or ((len(set(one_position)) == 1) and ('0/0' in set(one_position))) or ((len(set(one_position)) == 2) and ('0/0' in set(one_position) and ('./.' in set(one_position))))):

  216. 			pass

  217. 		else:

  218. 			consensus_output = row._1 + '\t' + str(row.POS) + '\t' + '.' + '\t' + row.REF + '\t' + row.ALT + '\t' + '.' + '\t' + '.' + '\t' +'.' + '\t' + 'GT' + '\t' + LCL5_consensus_call + '\t' + LCL6_consensus_call + '\t' + LCL7_consensus_call + '\t' + LCL8_consensus_call +'\n'

  219. 			consensus_outfile.write(consensus_output)

  220. 

  221. 

  222. 

  223. 

  224. 

  225. 

  226.