|
15
|
1 #!/usr/bin/python
|
|
|
2
|
|
|
3 import re
|
|
|
4 import sys
|
|
|
5 import optparse
|
|
|
6 import csv
|
|
|
7 import re
|
|
|
8 import pprint
|
|
|
9 from decimal import *
|
|
|
10 from rpy import *
|
|
|
11
|
|
|
12 def main():
|
|
|
13 csv.field_size_limit(1000000000)
|
|
|
14
|
|
|
15 parser = optparse.OptionParser()
|
|
|
16 parser.add_option('-p', '--sample_pileup', dest = 'sample_pileup', action = 'store', type = 'string', default = None, help = "Sample pileup from mpileup")
|
|
|
17 parser.add_option('-v', '--haw_vcf', dest = 'haw_vcf', action = 'store', type = 'string', default = None, help = "vcf file of Hawaiian SNPs")
|
|
|
18 parser.add_option('-l', '--loess_span', dest = 'loess_span', action = 'store', type = 'float', default = .01, help = "Loess span")
|
|
|
19 parser.add_option('-d', '--d_yaxis', dest = 'd_yaxis', action = 'store', type = 'float', default = .7, help = "y-axis upper limit for dot plot")
|
|
|
20 parser.add_option('-y', '--h_yaxis', dest = 'h_yaxis', action = 'store', type = 'int', default = 5, help = "y-axis upper limit for histogram plot")
|
|
|
21 parser.add_option('-c', '--points_color', dest = 'points_color', action = 'store', type = 'string', default = "gray27", help = "Color for data points")
|
|
|
22 parser.add_option('-k', '--loess_color', dest = 'loess_color', action = 'store', type = 'string', default = "red", help = "Color for loess regression line")
|
|
|
23 parser.add_option('-z', '--standardize', dest = 'standardize', default= 'true', help = "Standardize X-axis")
|
|
|
24 parser.add_option('-b', '--break_file', dest = 'break_file', action = 'store', type = 'string', default = 'C.elegans', help = "File defining the breaks per chromosome")
|
|
|
25 parser.add_option('-x', '--bin_size', dest = 'bin_size', action = 'store', type = 'int', default = 1000000, help = "Size of histogram bins, default is 1mb")
|
|
|
26 #parser.add_option('-n', '--normalize_bins', dest = 'normalize_bins', action = 'store_true', help = "Normalize histograms")
|
|
|
27 parser.add_option('-n', '--normalize_bins', dest = 'normalize_bins', default= 'true', help = "Normalize histograms")
|
|
|
28
|
|
|
29
|
|
|
30 parser.add_option('-o', '--output', dest = 'output', action = 'store', type = 'string', default = None, help = "Output file name")
|
|
|
31 parser.add_option('-s', '--location_plot_output', dest = 'location_plot_output', action = 'store', type = 'string', default = "SNP_Mapping_Plot.pdf", help = "Output file name of SNP plots by chromosomal location")
|
|
|
32
|
|
|
33 #For plotting with map units on the X-axis instead of physical distance
|
|
|
34 #parser.add_option('-u', '--mpu_plot_output', dest = 'mpu_plot_output', action = 'store', type = 'string', default = None, help = "Output file name of SNP plots by map unit location")
|
|
|
35 (options, args) = parser.parse_args()
|
|
|
36
|
|
|
37 haw_snps = build_haw_snp_dictionary(haw_vcf = options.haw_vcf)
|
|
|
38 pileup_info = parse_pileup(sample_pileup = options.sample_pileup, haw_snps = haw_snps)
|
|
|
39
|
|
|
40 output_pileup_info(output = options.output, pileup_info = pileup_info)
|
|
|
41
|
|
|
42 #output plot with all ratios
|
|
|
43 rounded_bin_size = int(round((float(options.bin_size) / 1000000), 1) * 1000000)
|
|
|
44
|
|
|
45 normalized_histogram_bins_per_mb = calculate_normalized_histogram_bins_per_xbase(pileup_info = pileup_info, xbase = rounded_bin_size, normalize_bins = options.normalize_bins)
|
|
|
46 normalized_histogram_bins_per_5kb = calculate_normalized_histogram_bins_per_xbase(pileup_info = pileup_info, xbase = (rounded_bin_size / 2), normalize_bins = options.normalize_bins)
|
|
|
47
|
|
|
48 break_dict = parse_breaks(break_file = options.break_file)
|
|
|
49
|
|
|
50 output_scatter_plots_by_location(location_plot_output = options.location_plot_output, pileup_info = pileup_info, loess_span=options.loess_span, d_yaxis=options.d_yaxis, h_yaxis=options.h_yaxis, points_color=options.points_color, loess_color=options.loess_color, standardize =options.standardize, normalized_hist_per_mb = normalized_histogram_bins_per_mb, normalized_hist_per_5kb = normalized_histogram_bins_per_5kb, breaks = break_dict, rounded_bin_size = rounded_bin_size)
|
|
|
51
|
|
|
52 #For plotting with map units on the X-axis instead of physical distance)
|
|
|
53 #output_scatter_plots_by_mapping_units(mpu_plot_output = options.mpu_plot_output, pileup_info = pileup_info)
|
|
|
54
|
|
|
55 def skip_headers(reader = None, i_file = None):
|
|
|
56 # count headers
|
|
|
57 comment = 0
|
|
|
58 while reader.next()[0].startswith('#'):
|
|
|
59 comment = comment + 1
|
|
|
60
|
|
|
61 # skip headers
|
|
|
62 i_file.seek(0)
|
|
|
63 for i in range(0, comment):
|
|
|
64 reader.next()
|
|
|
65
|
|
|
66 def parse_breaks(break_file = None):
|
|
|
67 if break_file == 'C.elegans':
|
|
|
68 break_dict = { 'I' : 16 , 'II' : 16, 'III' : 14, 'IV' : 18, 'V' : 21, 'X' : 18 }
|
|
|
69 return break_dict
|
|
|
70 elif break_file == 'Arabadopsis':
|
|
|
71 break_dict = { '1' : 16 , '2' : 16, '3' : 21, '4' : 18, '5' : 21 }
|
|
|
72 return break_dict
|
|
|
73 else:
|
|
|
74 i_file = open(break_file, 'rU')
|
|
|
75 break_dict = {}
|
|
|
76 reader = csv.reader(i_file, delimiter = '\t')
|
|
|
77 for row in reader:
|
|
|
78 chromosome = row[0].upper()
|
|
|
79 chromosome = re.sub("chr", "", chromosome, flags = re.IGNORECASE)
|
|
|
80 chromosome = re.sub("CHROMOSOME_", "", chromosome, flags = re.IGNORECASE)
|
|
|
81 break_count = row[1]
|
|
|
82 break_dict[chromosome] = int(break_count)
|
|
|
83 return break_dict
|
|
|
84
|
|
|
85
|
|
|
86 def location_comparer(location_1, location_2):
|
|
|
87 chr_loc_1 = location_1.split(':')[0]
|
|
|
88 pos_loc_1 = int(location_1.split(':')[1])
|
|
|
89
|
|
|
90 chr_loc_2 = location_2.split(':')[0]
|
|
|
91 pos_loc_2 = int(location_2.split(':')[1])
|
|
|
92
|
|
|
93 if chr_loc_1 == chr_loc_2:
|
|
|
94 if pos_loc_1 < pos_loc_2:
|
|
|
95 return -1
|
|
|
96 elif pos_loc_1 == pos_loc_1:
|
|
|
97 return 0
|
|
|
98 elif pos_loc_1 > pos_loc_2:
|
|
|
99 return 1
|
|
|
100 elif chr_loc_1 < chr_loc_2:
|
|
|
101 return -1
|
|
|
102 elif chr_loc_1 > chr_loc_2:
|
|
|
103 return 1
|
|
|
104
|
|
|
105 def output_pileup_info(output = None, pileup_info = None):
|
|
|
106 o_file = open(output, 'wb')
|
|
|
107 writer = csv.writer(o_file, delimiter = '\t')
|
|
|
108
|
|
|
109 writer.writerow(["#Chr\t", "Pos\t", "ID\t", "Alt Count\t", "Ref Count\t", "Read Depth\t", "Ratio\t", "Mapping Unit"])
|
|
|
110
|
|
|
111 location_sorted_pileup_info_keys = sorted(pileup_info.keys(), cmp=location_comparer)
|
|
|
112
|
|
|
113 for location in location_sorted_pileup_info_keys:
|
|
|
114 alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit = pileup_info[location]
|
|
|
115
|
|
|
116 location_info = location.split(':')
|
|
|
117 chromosome = location_info[0]
|
|
|
118 position = location_info[1]
|
|
|
119
|
|
|
120 writer.writerow([chromosome, position, haw_snp_id, alt_allele_count, ref_allele_count, read_depth, ratio, mapping_unit])
|
|
|
121
|
|
|
122 o_file.close()
|
|
|
123
|
|
|
124 def output_scatter_plots_by_location(location_plot_output = None, pileup_info = None, loess_span="", d_yaxis="", h_yaxis="", points_color="", loess_color="", standardize=None, normalized_hist_per_mb = None, normalized_hist_per_5kb = None, breaks = None, rounded_bin_size = 1000000):
|
|
|
125 positions = {}
|
|
|
126 current_chr = ""
|
|
|
127 prev_chr = ""
|
|
|
128
|
|
|
129 x_label = "Location (Mb)"
|
|
|
130 filtered_label = ''
|
|
|
131
|
|
|
132 location_sorted_pileup_info_keys = sorted(pileup_info.keys(), cmp=location_comparer)
|
|
|
133
|
|
|
134 break_unit = Decimal(rounded_bin_size) / Decimal(1000000)
|
|
|
135 max_breaks = max(breaks.values())
|
|
|
136
|
|
|
137 try:
|
|
|
138 r.pdf(location_plot_output, 8, 8)
|
|
|
139
|
|
|
140 for location in location_sorted_pileup_info_keys:
|
|
|
141 current_chr = location.split(':')[0]
|
|
|
142 position = location.split(':')[1]
|
|
|
143
|
|
|
144 alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit = pileup_info[location]
|
|
|
145
|
|
|
146 if prev_chr != current_chr:
|
|
|
147 if prev_chr != "":
|
|
|
148 hist_dict_mb = get_hist_dict_by_chr(normalized_hist_per_xbase = normalized_hist_per_mb, chr = prev_chr)
|
|
|
149 hist_dict_5kb = get_hist_dict_by_chr(normalized_hist_per_xbase = normalized_hist_per_5kb, chr = prev_chr)
|
|
|
150
|
|
|
151 plot_data(chr_dict = positions, hist_dict_mb = hist_dict_mb, hist_dict_5kb = hist_dict_5kb, chr = prev_chr + filtered_label, x_label = "Location (Mb)", divide_position = True, draw_secondary_grid_lines = True, loess_span=loess_span, d_yaxis=d_yaxis, h_yaxis=h_yaxis, points_color=points_color, loess_color=loess_color, breaks = breaks[prev_chr], standardize=standardize, max_breaks = max_breaks, break_unit = break_unit)
|
|
|
152
|
|
|
153 prev_chr = current_chr
|
|
|
154 positions = {}
|
|
|
155
|
|
|
156 positions[position] = ratio
|
|
|
157
|
|
|
158 hist_dict_mb = get_hist_dict_by_chr(normalized_hist_per_xbase = normalized_hist_per_mb, chr = current_chr)
|
|
|
159 hist_dict_5kb = get_hist_dict_by_chr(normalized_hist_per_xbase = normalized_hist_per_5kb, chr = current_chr)
|
|
|
160
|
|
|
161 plot_data(chr_dict = positions, hist_dict_mb = hist_dict_mb, hist_dict_5kb = hist_dict_5kb, chr = current_chr + filtered_label, x_label = "Location (Mb)", divide_position = True, draw_secondary_grid_lines = True, loess_span=loess_span, d_yaxis=d_yaxis, h_yaxis=h_yaxis, points_color=points_color, loess_color=loess_color, breaks = breaks[current_chr], standardize=standardize, max_breaks = max_breaks, break_unit = break_unit)
|
|
|
162
|
|
|
163 r.dev_off()
|
|
|
164
|
|
|
165 except Exception as inst:
|
|
|
166 print inst
|
|
|
167 print "There was an error creating the location plot pdf... Please try again"
|
|
|
168
|
|
|
169 def get_hist_dict_by_chr(normalized_hist_per_xbase = None, chr = ''):
|
|
|
170 hist_dict = {}
|
|
|
171
|
|
|
172 for location in normalized_hist_per_xbase:
|
|
|
173 chromosome = location.split(':')[0]
|
|
|
174 if chromosome == chr:
|
|
|
175 position = int(location.split(':')[1])
|
|
|
176 hist_dict[position] = normalized_hist_per_xbase[location]
|
|
|
177
|
|
|
178 return hist_dict
|
|
|
179
|
|
|
180 '''
|
|
|
181 def output_scatter_plots_by_mapping_units(mpu_plot_output = None, pileup_info = None):
|
|
|
182 i = {}
|
|
|
183 ii = {}
|
|
|
184 iii = {}
|
|
|
185 iv = {}
|
|
|
186 v = {}
|
|
|
187 x = {}
|
|
|
188
|
|
|
189 for location in pileup_info:
|
|
|
190 chromosome = location.split(':')[0]
|
|
|
191 position = location.split(':')[1]
|
|
|
192
|
|
|
193 alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit = pileup_info[location]
|
|
|
194
|
|
|
195 if chromosome == "I":
|
|
|
196 i[mapping_unit] = ratio
|
|
|
197 elif chromosome == "II":
|
|
|
198 ii[mapping_unit] = ratio
|
|
|
199 elif chromosome == "III":
|
|
|
200 iii[mapping_unit] = ratio
|
|
|
201 elif chromosome == "IV":
|
|
|
202 iv[mapping_unit] = ratio
|
|
|
203 elif chromosome == "V":
|
|
|
204 v[mapping_unit] = ratio
|
|
|
205 elif chromosome == "X":
|
|
|
206 x[mapping_unit] = ratio
|
|
|
207
|
|
|
208 x_label = "Map Units"
|
|
|
209
|
|
|
210 try:
|
|
|
211 r.pdf(mpu_plot_output, 8, 8)
|
|
|
212 plot_data(chr_dict = i, chr = "I", x_label = "Map Units")
|
|
|
213 plot_data(chr_dict = ii, chr = "II", x_label = "Map Units")
|
|
|
214 plot_data(chr_dict = iii, chr = "III", x_label = "Map Units")
|
|
|
215 plot_data(chr_dict = iv, chr = "IV", x_label = "Map Units")
|
|
|
216 plot_data(chr_dict = v, chr = "V", x_label = "Map Units")
|
|
|
217 plot_data(chr_dict = x, chr = "X", x_label = "Map Units")
|
|
|
218 r.dev_off()
|
|
|
219 except Exception as inst:
|
|
|
220 print inst
|
|
|
221 print "There was an error creating the map unit plot pdf... Please try again"
|
|
|
222 '''
|
|
|
223
|
|
|
224 def plot_data(chr_dict = None, hist_dict_mb = None, hist_dict_5kb = None, chr = "", x_label = "", divide_position = False, draw_secondary_grid_lines = False, loess_span=None, d_yaxis=None, h_yaxis=None, points_color="", loess_color="", breaks = None, standardize= None, max_breaks = 1, break_unit = 1):
|
|
|
225 ratios = "c("
|
|
|
226 positions = "c("
|
|
|
227
|
|
|
228 for position in chr_dict:
|
|
|
229 ratio = chr_dict[position]
|
|
|
230 if divide_position:
|
|
|
231 position = float(position) / 1000000.0
|
|
|
232 positions = positions + str(position) + ", "
|
|
|
233 ratios = ratios + str(ratio) + ", "
|
|
|
234
|
|
|
235 if len(ratios) == 2:
|
|
|
236 ratios = ratios + ")"
|
|
|
237 else:
|
|
|
238 ratios = ratios[0:len(ratios) - 2] + ")"
|
|
|
239
|
|
|
240 if len(positions) == 2:
|
|
|
241 positions = positions + ")"
|
|
|
242 else:
|
|
|
243 positions = positions[0:len(positions) - 2] + ")"
|
|
|
244
|
|
|
245 r("x <- " + positions)
|
|
|
246 r("y <- " + ratios)
|
|
|
247
|
|
|
248 hist_mb_values = "c("
|
|
|
249 for position in sorted(hist_dict_mb):
|
|
|
250 hist_mb_values = hist_mb_values + str(hist_dict_mb[position]) + ", "
|
|
|
251
|
|
|
252 if len(hist_mb_values) == 2:
|
|
|
253 hist_mb_values = hist_mb_values + ")"
|
|
|
254 else:
|
|
|
255 hist_mb_values = hist_mb_values[0:len(hist_mb_values) - 2] + ")"
|
|
|
256
|
|
|
257 hist_5kb_values = "c("
|
|
|
258 for position in sorted(hist_dict_5kb):
|
|
|
259 hist_5kb_values = hist_5kb_values + str(hist_dict_5kb[position]) + ", "
|
|
|
260
|
|
|
261 if len(hist_5kb_values) == 2:
|
|
|
262 hist_5kb_values = hist_5kb_values + ")"
|
|
|
263 else:
|
|
|
264 hist_5kb_values = hist_5kb_values[0:len(hist_5kb_values) - 2] + ")"
|
|
|
265
|
|
|
266 r("xz <- " + hist_mb_values)
|
|
|
267 r("yz <- " + hist_5kb_values)
|
|
|
268
|
|
|
269 max_break_str = str(max_breaks)
|
|
|
270 break_unit_str = str(Decimal(break_unit))
|
|
|
271 half_break_unit_str = str(Decimal(break_unit) / Decimal(2))
|
|
|
272 break_penta_unit_str = str(Decimal(break_unit) * Decimal(5))
|
|
|
273
|
|
|
274 if (standardize=='true'):
|
|
|
275 r("plot(x, y, ,cex=0.60, xlim=c(0," + max_break_str + "), main='LG " + chr + "', xlab= '" + x_label + "', ylim = c(0, %f " %d_yaxis + "), ylab='Ratios of mapping strain alleles/total reads (at SNP positions)', pch=18, col='"+ points_color +"')")
|
|
|
276 r("lines(loess.smooth(x, y, span = %f "%loess_span + "), lwd=5, col='"+ loess_color +"')")
|
|
|
277 r("axis(1, at=seq(0, " + max_break_str + ", by=" + break_unit_str + "), labels=FALSE, tcl=-0.5)")
|
|
|
278 r("axis(1, at=seq(0, " + max_break_str + ", by=" + half_break_unit_str + "), labels=FALSE, tcl=-0.25)")
|
|
|
279 r("axis(2, at=seq(floor(min(y)), 1, by=0.1), labels=FALSE, tcl=-0.2)")
|
|
|
280 elif (standardize=='false'):
|
|
|
281 r("plot(x, y, cex=0.60, main='LG " + chr + "', xlab= '" + x_label + "', ylim = c(0, %f " %d_yaxis + "), ylab='Ratios of mapping strain alleles/total reads (at SNP positions)', pch=18, col='"+ points_color +"')")
|
|
|
282 r("lines(loess.smooth(x, y, span = %f "%loess_span + "), lwd=5, col='"+ loess_color +"')")
|
|
|
283 r("axis(1, at=seq(0, as.integer( ' " + str(breaks) + " '), by= " + break_unit_str + "), labels=FALSE, tcl=-0.5)")
|
|
|
284 r("axis(1, at=seq(0, as.integer( ' " + str(breaks) + " '), by= " + half_break_unit_str + "), labels=FALSE, tcl=-0.25)")
|
|
|
285 r("axis(2, at=seq(floor(min(y)), 1, by=0.1), labels=FALSE, tcl=-0.2)")
|
|
|
286
|
|
|
287 if draw_secondary_grid_lines:
|
|
|
288 r("abline(h = seq(floor(min(y)), 1, by=0.1), v = seq(floor(min(x)), length(x), by= 1), col='gray')")
|
|
|
289 else:
|
|
|
290 r("grid(lty = 1, col = 'gray')")
|
|
|
291
|
|
|
292 if (standardize=='true'):
|
|
|
293 r("barplot(xz, xlim=c(0, " + max_break_str + "), ylim = c(0, " + str(h_yaxis) + "), yaxp=c(0, " + str(h_yaxis) + ", 1), space = 0, col='darkgray', width = " + break_unit_str + ", xlab='Location (Mb)', ylab='Normalized frequency of pure parental alleles ', main='LG " + chr + "')")
|
|
|
294 r("barplot(yz, space = 0, add=TRUE, width = " + half_break_unit_str + ", col=rgb(1, 0, 0, 1))")
|
|
|
295 r("axis(1, hadj = 1, at=seq(0, " + max_break_str + ", by= " + break_unit_str + "), labels=FALSE, tcl=-0.5)")
|
|
|
296 r("axis(1, at=seq(0, " + max_break_str + ", by= " + break_penta_unit_str + "), labels=TRUE, tcl=-0.5)")
|
|
|
297 r("axis(1, at=seq(0, " + max_break_str + ", by= " + half_break_unit_str + "), labels=FALSE, tcl=-0.25)")
|
|
|
298 elif (standardize=='false'):
|
|
|
299 r("barplot(xz, ylim = c(0, " + str(h_yaxis) + "), yaxp=c(0, " + str(h_yaxis) + ", 1), space = 0, col='darkgray', width = 1, xlab='Location (Mb)', ylab='Normalized frequency of pure parental alleles ', main='LG " + chr + "')")
|
|
|
300 r("barplot(yz, space = 0, add=TRUE, width = 0.5, col=rgb(1, 0, 0, 1))")
|
|
|
301 r("axis(1, at=seq(0, as.integer( ' " + str(breaks) + " '), by= " + break_unit_str + "), labels=FALSE, tcl=-0.5)")
|
|
|
302 r("axis(1, at=seq(0, as.integer( ' " + str(breaks) + " '), by= " + break_penta_unit_str + "), labels=TRUE, tcl=-0.5)")
|
|
|
303 r("axis(1, at=seq(0, as.integer( ' " + str(breaks) + " '), by= " + half_break_unit_str + "), labels=FALSE, tcl=-0.25)")
|
|
|
304
|
|
|
305
|
|
|
306 def build_haw_snp_dictionary(haw_vcf = None):
|
|
|
307 haw_snps = {}
|
|
|
308
|
|
|
309 i_file = open(haw_vcf, 'rU')
|
|
|
310 reader = csv.reader(i_file, delimiter = '\t')
|
|
|
311
|
|
|
312 skip_headers(reader = reader, i_file = i_file)
|
|
|
313
|
|
|
314 for row in reader:
|
|
|
315 #print row
|
|
|
316 chromosome = row[0].upper()
|
|
|
317 chromosome = re.sub("chr", "", chromosome, flags = re.IGNORECASE)
|
|
|
318 chromosome = re.sub("CHROMOSOME_", "", chromosome, flags = re.IGNORECASE)
|
|
|
319
|
|
|
320 position = row[1]
|
|
|
321 haw_snp_id = row[2]
|
|
|
322 ref_allele = row[3]
|
|
|
323 alt_allele = row[4]
|
|
|
324
|
|
|
325 info = row[7]
|
|
|
326
|
|
|
327 mapping_unit = info.replace("MPU=", "")
|
|
|
328
|
|
|
329 location = chromosome + ":" + position
|
|
|
330 haw_snps[location] = (alt_allele, haw_snp_id, mapping_unit)
|
|
|
331
|
|
|
332 i_file.close()
|
|
|
333
|
|
|
334 return haw_snps
|
|
|
335
|
|
|
336 def calculate_normalized_histogram_bins_per_xbase(pileup_info = None, xbase = 1000000, normalize_bins = None):
|
|
|
337 normalized_histogram_bins_per_xbase = {}
|
|
|
338
|
|
|
339 ref_snp_count_per_xbase = get_ref_snp_count_per_xbase(pileup_info = pileup_info, xbase = xbase)
|
|
|
340 mean_zero_snp_count_per_chromosome = get_mean_zero_snp_count_per_chromosome(pileup_info = pileup_info, xbase = xbase)
|
|
|
341 zero_snp_count_per_xbase = get_zero_snp_count_per_xbase(pileup_info = pileup_info, xbase = xbase)
|
|
|
342
|
|
|
343 for location in ref_snp_count_per_xbase:
|
|
|
344 chromosome = location.split(':')[0]
|
|
|
345 mean_zero_snp_count = mean_zero_snp_count_per_chromosome[chromosome]
|
|
|
346 ref_snp_count = ref_snp_count_per_xbase[location]
|
|
|
347
|
|
|
348 zero_snp_count = 0
|
|
|
349 if location in zero_snp_count_per_xbase:
|
|
|
350 zero_snp_count = zero_snp_count_per_xbase[location]
|
|
|
351
|
|
|
352 if normalize_bins == 'true':
|
|
|
353 if zero_snp_count == 0 or ref_snp_count == 0:
|
|
|
354 normalized_histogram_bins_per_xbase[location] = 0
|
|
|
355 elif zero_snp_count == ref_snp_count:
|
|
|
356 normalized_histogram_bins_per_xbase[location] = 0
|
|
|
357 else:
|
|
|
358 normalized_histogram_bins_per_xbase[location] = (Decimal(zero_snp_count) / (Decimal(ref_snp_count)-Decimal(zero_snp_count))) * Decimal(mean_zero_snp_count)
|
|
|
359 else:
|
|
|
360 normalized_histogram_bins_per_xbase[location] = zero_snp_count
|
|
|
361
|
|
|
362 return normalized_histogram_bins_per_xbase
|
|
|
363
|
|
|
364 def get_ref_snp_count_per_xbase(pileup_info = None, xbase = 1000000):
|
|
|
365 ref_snps_per_xbase = {}
|
|
|
366
|
|
|
367 for location in pileup_info:
|
|
|
368 location_info = location.split(':')
|
|
|
369
|
|
|
370 chromosome = location_info[0].upper()
|
|
|
371 chromosome = re.sub("chr", "", chromosome, flags = re.IGNORECASE)
|
|
|
372 chromosome = re.sub("CHROMOSOME_", "", chromosome, flags = re.IGNORECASE)
|
|
|
373
|
|
|
374 position = location_info[1]
|
|
|
375 xbase_position = (int(position) / xbase) + 1
|
|
|
376
|
|
|
377 location = chromosome + ":" + str(xbase_position)
|
|
|
378 if location in ref_snps_per_xbase:
|
|
|
379 ref_snps_per_xbase[location] = ref_snps_per_xbase[location] + 1
|
|
|
380 else:
|
|
|
381 ref_snps_per_xbase[location] = 1
|
|
|
382
|
|
|
383 return ref_snps_per_xbase
|
|
|
384
|
|
|
385 def get_mean_zero_snp_count_per_chromosome(pileup_info, xbase = 1000000):
|
|
|
386 sample_snp_count_per_xbase = {}
|
|
|
387
|
|
|
388 for location in pileup_info:
|
|
|
389 alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit = pileup_info[location]
|
|
|
390
|
|
|
391 location_info = location.split(':')
|
|
|
392 chromosome = location_info[0]
|
|
|
393 position = location_info[1]
|
|
|
394 xbase_position = (int(position) / xbase) + 1
|
|
|
395 xbase_location = chromosome + ":" + str(xbase_position)
|
|
|
396
|
|
|
397 if alt_allele_count == 0:
|
|
|
398 if xbase_location in sample_snp_count_per_xbase:
|
|
|
399 sample_snp_count_per_xbase[xbase_location] = sample_snp_count_per_xbase[xbase_location] + 1
|
|
|
400 else:
|
|
|
401 sample_snp_count_per_xbase[xbase_location] = 1
|
|
|
402
|
|
|
403 elif alt_allele_count != 0 and xbase_location not in sample_snp_count_per_xbase:
|
|
|
404 sample_snp_count_per_xbase[xbase_location] = 0
|
|
|
405
|
|
|
406 mean_zero_snp_count_per_chromosome = {}
|
|
|
407 for location in sample_snp_count_per_xbase:
|
|
|
408 chromosome = location.split(':')[0]
|
|
|
409 sample_count = sample_snp_count_per_xbase[location]
|
|
|
410 if chromosome in mean_zero_snp_count_per_chromosome:
|
|
|
411 mean_zero_snp_count_per_chromosome[chromosome].append(sample_count)
|
|
|
412 else:
|
|
|
413 mean_zero_snp_count_per_chromosome[chromosome] = [sample_count]
|
|
|
414
|
|
|
415 for chromosome in mean_zero_snp_count_per_chromosome:
|
|
|
416 summa = sum(mean_zero_snp_count_per_chromosome[chromosome])
|
|
|
417 count = len(mean_zero_snp_count_per_chromosome[chromosome])
|
|
|
418
|
|
|
419 mean_zero_snp_count_per_chromosome[chromosome] = Decimal(summa) / Decimal(count)
|
|
|
420
|
|
|
421 return mean_zero_snp_count_per_chromosome
|
|
|
422
|
|
|
423 def get_zero_snp_count_per_xbase(pileup_info = None, xbase = 1000000):
|
|
|
424 zero_snp_count_per_xbase = {}
|
|
|
425
|
|
|
426 for location in pileup_info:
|
|
|
427 alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit = pileup_info[location]
|
|
|
428
|
|
|
429 location_info = location.split(':')
|
|
|
430 chromosome = location_info[0]
|
|
|
431 position = location_info[1]
|
|
|
432 xbase_position = (int(position) / xbase) + 1
|
|
|
433 xbase_location = chromosome + ":" + str(xbase_position)
|
|
|
434
|
|
|
435 if alt_allele_count == 0:
|
|
|
436 if xbase_location in zero_snp_count_per_xbase:
|
|
|
437 zero_snp_count_per_xbase[xbase_location] = zero_snp_count_per_xbase[xbase_location] + 1
|
|
|
438 else:
|
|
|
439 zero_snp_count_per_xbase[xbase_location] = 1
|
|
|
440
|
|
|
441 elif alt_allele_count != 0 and xbase_location not in zero_snp_count_per_xbase:
|
|
|
442 zero_snp_count_per_xbase[xbase_location] = 0
|
|
|
443
|
|
|
444 return zero_snp_count_per_xbase
|
|
|
445
|
|
|
446 def parse_pileup(sample_pileup = None, haw_snps = None):
|
|
|
447 i_file = open(sample_pileup, 'rU')
|
|
|
448 reader = csv.reader(i_file, delimiter = '\t', quoting = csv.QUOTE_NONE)
|
|
|
449
|
|
|
450 pileup_info = {}
|
|
|
451
|
|
|
452 for row in reader:
|
|
|
453 chromosome = row[0].upper()
|
|
|
454 chromosome = re.sub("chr", "", chromosome, flags = re.IGNORECASE)
|
|
|
455 chromosome = re.sub("CHROMOSOME_", "", chromosome, flags = re.IGNORECASE)
|
|
|
456
|
|
|
457 position = row[1]
|
|
|
458 ref_allele = row[2]
|
|
|
459 read_depth = row[3]
|
|
|
460 read_bases = row[4]
|
|
|
461
|
|
|
462 location = chromosome + ":" + position
|
|
|
463 if location in haw_snps:
|
|
|
464 alt_allele, haw_snp_id, mapping_unit = haw_snps[location]
|
|
|
465 ref_allele_count, alt_allele_count = parse_read_bases(read_bases = read_bases, alt_allele = alt_allele)
|
|
|
466
|
|
|
467 if Decimal(read_depth!=0):
|
|
|
468 getcontext().prec = 6
|
|
|
469 ratio = Decimal(alt_allele_count) / Decimal(read_depth)
|
|
|
470 #ratio = Decimal(alt_allele_count) / Decimal(ref_allele_count)
|
|
|
471
|
|
|
472 pileup_info[location] = (alt_allele_count, ref_allele_count, read_depth, ratio, haw_snp_id, mapping_unit)
|
|
|
473
|
|
|
474 #debug line
|
|
|
475 #print chromosome, position, read_depth, ref_allele_count, alt_allele_count, ratio, id
|
|
|
476
|
|
|
477 i_file.close()
|
|
|
478
|
|
|
479 return pileup_info
|
|
|
480
|
|
|
481 def parse_read_bases(read_bases = None, alt_allele = None):
|
|
|
482 read_bases = re.sub('\$', '', read_bases)
|
|
|
483 read_bases = re.sub('\^[^\s]', '', read_bases)
|
|
|
484
|
|
|
485 ref_allele_matches = re.findall("\.|\,", read_bases)
|
|
|
486 ref_allele_count = len(ref_allele_matches)
|
|
|
487
|
|
|
488 alt_allele_matches = re.findall(alt_allele, read_bases, flags = re.IGNORECASE)
|
|
|
489 alt_allele_count = len(alt_allele_matches)
|
|
|
490
|
|
|
491 #debug line
|
|
|
492 #print read_bases, alt_allele, alt_allele_count, ref_allele_count
|
|
|
493
|
|
|
494 return ref_allele_count, alt_allele_count
|
|
|
495
|
|
|
496 if __name__ == "__main__":
|
|
|
497 main()
|
