1 | # utilities for rgenetics |
---|
2 | # |
---|
3 | # copyright 2009 ross lazarus |
---|
4 | # released under the LGPL |
---|
5 | # |
---|
6 | |
---|
7 | import subprocess, os, sys, time, tempfile,string,plinkbinJZ |
---|
8 | |
---|
9 | galhtmlprefix = """<?xml version="1.0" encoding="utf-8" ?> |
---|
10 | <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> |
---|
11 | <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"> |
---|
12 | <head> |
---|
13 | <meta http-equiv="Content-Type" content="text/html; charset=utf-8" /> |
---|
14 | <meta name="generator" content="Galaxy %s tool output - see http://g2.trac.bx.psu.edu/" /> |
---|
15 | <title></title> |
---|
16 | <link rel="stylesheet" href="/static/style/base.css" type="text/css" /> |
---|
17 | </head> |
---|
18 | <body> |
---|
19 | <div class="document"> |
---|
20 | """ |
---|
21 | galhtmlattr = """<h3><a href="http://rgenetics.org">Rgenetics</a> tool %s run at %s</h3>""" |
---|
22 | galhtmlpostfix = """</div></body></html>\n""" |
---|
23 | |
---|
24 | plinke = 'plink' # changed jan 2010 - all exes must be on path |
---|
25 | rexe = 'R' # to avoid cluster/platform dependencies |
---|
26 | smartpca = 'smartpca.perl' |
---|
27 | |
---|
28 | def timenow(): |
---|
29 | """return current time as a string |
---|
30 | """ |
---|
31 | return time.strftime('%d/%m/%Y %H:%M:%S', time.localtime(time.time())) |
---|
32 | |
---|
33 | def fail( message ): |
---|
34 | print >> sys.stderr, message |
---|
35 | return -1 |
---|
36 | |
---|
37 | def whereis(program): |
---|
38 | for path in os.environ.get('PATH', '').split(':'): |
---|
39 | if os.path.exists(os.path.join(path, program)) and \ |
---|
40 | not os.path.isdir(os.path.join(path, program)): |
---|
41 | return os.path.join(path, program) |
---|
42 | return None |
---|
43 | |
---|
44 | |
---|
45 | def oRRun(rcmd=[],outdir=None,title='myR',rexe='R'): |
---|
46 | """ |
---|
47 | run an r script, lines in rcmd, |
---|
48 | in a temporary directory |
---|
49 | move everything, r script and all back to outdir which will be an html file |
---|
50 | |
---|
51 | |
---|
52 | # test |
---|
53 | RRun(rcmd=['print("hello cruel world")','q()'],title='test') |
---|
54 | """ |
---|
55 | rlog = [] |
---|
56 | print '### rexe = %s' % rexe |
---|
57 | assert os.path.isfile(rexe) |
---|
58 | rname = '%s.R' % title |
---|
59 | stoname = '%s.R.log' % title |
---|
60 | rfname = rname |
---|
61 | stofname = stoname |
---|
62 | if outdir: # want a specific path |
---|
63 | rfname = os.path.join(outdir,rname) |
---|
64 | stofname = os.path.join(outdir,stoname) |
---|
65 | try: |
---|
66 | os.makedirs(outdir) # might not be there yet... |
---|
67 | except: |
---|
68 | pass |
---|
69 | else: |
---|
70 | outdir = tempfile.mkdtemp(prefix=title) |
---|
71 | rfname = os.path.join(outdir,rname) |
---|
72 | stofname = os.path.join(outdir,stoname) |
---|
73 | rmoutdir = True |
---|
74 | f = open(rfname,'w') |
---|
75 | if type(rcmd) == type([]): |
---|
76 | f.write('\n'.join(rcmd)) |
---|
77 | else: # string |
---|
78 | f.write(rcmd) |
---|
79 | f.write('\n') |
---|
80 | f.close() |
---|
81 | sto = file(stofname,'w') |
---|
82 | vcl = [rexe,"--vanilla --slave", '<', rfname ] |
---|
83 | x = subprocess.Popen(' '.join(vcl),shell=True,stderr=sto,stdout=sto,cwd=outdir) |
---|
84 | retval = x.wait() |
---|
85 | sto.close() |
---|
86 | rlog = file(stofname,'r').readlines() |
---|
87 | rlog.insert(0,'## found R at %s' % rexe) |
---|
88 | if outdir <> None: |
---|
89 | flist = os.listdir(outdir) |
---|
90 | else: |
---|
91 | flist = os.listdir('.') |
---|
92 | flist.sort |
---|
93 | flist = [(x,x) for x in flist] |
---|
94 | for i,x in enumerate(flist): |
---|
95 | if x == rname: |
---|
96 | flist[i] = (x,'R script for %s' % title) |
---|
97 | elif x == stoname: |
---|
98 | flist[i] = (x,'R log for %s' % title) |
---|
99 | if False and rmoutdir: |
---|
100 | os.removedirs(outdir) |
---|
101 | return rlog,flist # for html layout |
---|
102 | |
---|
103 | |
---|
104 | |
---|
105 | |
---|
106 | def RRun(rcmd=[],outdir=None,title='myR',tidy=True): |
---|
107 | """ |
---|
108 | run an r script, lines in rcmd, |
---|
109 | in a temporary directory |
---|
110 | move everything, r script and all back to outdir which will be an html file |
---|
111 | |
---|
112 | |
---|
113 | # test |
---|
114 | RRun(rcmd=['print("hello cruel world")','q()'],title='test') |
---|
115 | echo "a <- c(5, 5); b <- c(0.5, 0.5)" | cat - RScript.R | R --slave \ --vanilla |
---|
116 | suggested by http://tolstoy.newcastle.edu.au/R/devel/05/09/2448.html |
---|
117 | """ |
---|
118 | killme = string.punctuation + string.whitespace |
---|
119 | trantab = string.maketrans(killme,'_'*len(killme)) |
---|
120 | title = title.translate(trantab) |
---|
121 | rlog = [] |
---|
122 | tempout=False |
---|
123 | rname = '%s.R' % title |
---|
124 | stoname = '%s.R.log' % title |
---|
125 | cwd = os.getcwd() |
---|
126 | if outdir: # want a specific path |
---|
127 | try: |
---|
128 | os.makedirs(outdir) # might not be there yet... |
---|
129 | except: |
---|
130 | pass |
---|
131 | os.chdir(outdir) |
---|
132 | if type(rcmd) == type([]): |
---|
133 | script = '\n'.join(rcmd) |
---|
134 | else: # string |
---|
135 | script = rcmd |
---|
136 | sto = file(stoname,'w') |
---|
137 | rscript = file(rname,'w') |
---|
138 | rscript.write(script) |
---|
139 | rscript.write('\n#R script autogenerated by rgenetics/rgutils.py on %s\n' % timenow()) |
---|
140 | rscript.close() |
---|
141 | vcl = '%s --slave --vanilla < %s' % (rexe,rname) |
---|
142 | if outdir: |
---|
143 | x = subprocess.Popen(vcl,shell=True,stderr=sto,stdout=sto,cwd=outdir) |
---|
144 | else: |
---|
145 | x = subprocess.Popen(vcl,shell=True,stderr=sto,stdout=sto) |
---|
146 | retval = x.wait() |
---|
147 | sto.close() |
---|
148 | rlog = file(stoname,'r').readlines() |
---|
149 | if retval <> 0: |
---|
150 | rlog.insert(0,'Nonzero exit code = %d' % retval) # indicate failure |
---|
151 | if outdir: |
---|
152 | flist = os.listdir(outdir) |
---|
153 | else: |
---|
154 | flist = os.listdir(os.getcwd()) |
---|
155 | flist.sort |
---|
156 | flist = [(x,x) for x in flist] |
---|
157 | for i,x in enumerate(flist): |
---|
158 | if x == rname: |
---|
159 | flist[i] = (x,'R script for %s' % title) |
---|
160 | elif x == stoname: |
---|
161 | flist[i] = (x,'R log for %s' % title) |
---|
162 | if outdir: |
---|
163 | os.chdir(cwd) |
---|
164 | return rlog,flist # for html layout |
---|
165 | |
---|
166 | def runPlink(bfn='bar',ofn='foo',logf=None,plinktasks=[],cd='./',vclbase = []): |
---|
167 | """run a series of plink tasks and append log results to stdout |
---|
168 | vcl has a list of parameters for the spawnv |
---|
169 | common settings can all go in the vclbase list and are added to each plinktask |
---|
170 | """ |
---|
171 | # root for all |
---|
172 | fplog,plog = tempfile.mkstemp() |
---|
173 | if type(logf) == type(' '): # open otherwise assume is file - ugh I'm in a hurry |
---|
174 | mylog = file(logf,'a+') |
---|
175 | else: |
---|
176 | mylog = logf |
---|
177 | mylog.write('## Rgenetics: http://rgenetics.org Galaxy Tools rgQC.py Plink runner\n') |
---|
178 | for task in plinktasks: # each is a list |
---|
179 | vcl = vclbase + task |
---|
180 | sto = file(plog,'w') |
---|
181 | x = subprocess.Popen(' '.join(vcl),shell=True,stdout=sto,stderr=sto,cwd=cd) |
---|
182 | retval = x.wait() |
---|
183 | sto.close() |
---|
184 | try: |
---|
185 | lplog = file(plog,'r').read() |
---|
186 | mylog.write(lplog) |
---|
187 | os.unlink(plog) # no longer needed |
---|
188 | except: |
---|
189 | mylog.write('### %s Strange - no std out from plink when running command line\n%s' % (timenow(),' '.join(vcl))) |
---|
190 | |
---|
191 | def pruneLD(plinktasks=[],cd='./',vclbase = []): |
---|
192 | """ |
---|
193 | plink blathers when doing pruning - ignore |
---|
194 | Linkage disequilibrium based SNP pruning |
---|
195 | if a million snps in 3 billion base pairs, have mean 3k spacing |
---|
196 | assume 40-60k of ld in ceu, a window of 120k width is about 40 snps |
---|
197 | so lots more is perhaps less efficient - each window computational cost is |
---|
198 | ON^2 unless the code is smart enough to avoid unecessary computation where |
---|
199 | allele frequencies make it impossible to see ld > the r^2 cutoff threshold |
---|
200 | So, do a window and move forward 20? |
---|
201 | The fine Plink docs at http://pngu.mgh.harvard.edu/~purcell/plink/summary.shtml#prune |
---|
202 | reproduced below |
---|
203 | |
---|
204 | Sometimes it is useful to generate a pruned subset of SNPs that are in approximate linkage equilibrium with each other. This can be achieved via two commands: --indep which prunes based on the variance inflation factor (VIF), which recursively removes SNPs within a sliding window; second, --indep-pairwise which is similar, except it is based only on pairwise genotypic correlation. |
---|
205 | |
---|
206 | Hint The output of either of these commands is two lists of SNPs: those that are pruned out and those that are not. A separate command using the --extract or --exclude option is necessary to actually perform the pruning. |
---|
207 | |
---|
208 | The VIF pruning routine is performed: |
---|
209 | plink --file data --indep 50 5 2 |
---|
210 | |
---|
211 | will create files |
---|
212 | |
---|
213 | plink.prune.in |
---|
214 | plink.prune.out |
---|
215 | |
---|
216 | Each is a simlpe list of SNP IDs; both these files can subsequently be specified as the argument for |
---|
217 | a --extract or --exclude command. |
---|
218 | |
---|
219 | The parameters for --indep are: window size in SNPs (e.g. 50), the number of SNPs to shift the |
---|
220 | window at each step (e.g. 5), the VIF threshold. The VIF is 1/(1-R^2) where R^2 is the multiple correlation coefficient for a SNP being regressed on all other SNPs simultaneously. That is, this considers the correlations between SNPs but also between linear combinations of SNPs. A VIF of 10 is often taken to represent near collinearity problems in standard multiple regression analyses (i.e. implies R^2 of 0.9). A VIF of 1 would imply that the SNP is completely independent of all other SNPs. Practically, values between 1.5 and 2 should probably be used; particularly in small samples, if this threshold is too low and/or the window size is too large, too many SNPs may be removed. |
---|
221 | |
---|
222 | The second procedure is performed: |
---|
223 | plink --file data --indep-pairwise 50 5 0.5 |
---|
224 | |
---|
225 | This generates the same output files as the first version; the only difference is that a |
---|
226 | simple pairwise threshold is used. The first two parameters (50 and 5) are the same as above (window size and step); the third parameter represents the r^2 threshold. Note: this represents the pairwise SNP-SNP metric now, not the multiple correlation coefficient; also note, this is based on the genotypic correlation, i.e. it does not involve phasing. |
---|
227 | |
---|
228 | To give a concrete example: the command above that specifies 50 5 0.5 would a) consider a |
---|
229 | window of 50 SNPs, b) calculate LD between each pair of SNPs in the window, b) remove one of a pair of SNPs if the LD is greater than 0.5, c) shift the window 5 SNPs forward and repeat the procedure. |
---|
230 | |
---|
231 | To make a new, pruned file, then use something like (in this example, we also convert the |
---|
232 | standard PED fileset to a binary one): |
---|
233 | plink --file data --extract plink.prune.in --make-bed --out pruneddata |
---|
234 | """ |
---|
235 | fplog,plog = tempfile.mkstemp() |
---|
236 | alog = [] |
---|
237 | alog.append('## Rgenetics: http://rgenetics.org Galaxy Tools rgQC.py Plink pruneLD runner\n') |
---|
238 | for task in plinktasks: # each is a list |
---|
239 | vcl = vclbase + task |
---|
240 | sto = file(plog,'w') |
---|
241 | x = subprocess.Popen(' '.join(vcl),shell=True,stdout=sto,stderr=sto,cwd=cd) |
---|
242 | retval = x.wait() |
---|
243 | sto.close() |
---|
244 | try: |
---|
245 | lplog = file(plog,'r').readlines() |
---|
246 | lplog = [x for x in lplog if x.find('Pruning SNP') == -1] |
---|
247 | alog += lplog |
---|
248 | alog.append('\n') |
---|
249 | os.unlink(plog) # no longer needed |
---|
250 | except: |
---|
251 | alog.append('### %s Strange - no std out from plink when running command line\n%s\n' % (timenow(),' '.join(vcl))) |
---|
252 | return alog |
---|
253 | |
---|
254 | def readMap(mapfile=None,allmarkers=False,rsdict={},c=None,spos=None,epos=None): |
---|
255 | """abstract out - keeps reappearing |
---|
256 | """ |
---|
257 | mfile = open(mapfile, 'r') |
---|
258 | markers = [] |
---|
259 | snpcols = {} |
---|
260 | snpIndex = 0 # in case empty or comment lines |
---|
261 | for rownum,row in enumerate(mfile): |
---|
262 | line = row.strip() |
---|
263 | if not line or line[0]=='#': continue |
---|
264 | chrom, snp, genpos, abspos = line.split()[:4] # just in case more cols |
---|
265 | try: |
---|
266 | abspos = int(abspos) |
---|
267 | except: |
---|
268 | abspos = 0 # stupid framingham data grumble grumble |
---|
269 | if allmarkers or rsdict.get(snp,None) or (chrom == c and (spos <= abspos <= epos)): |
---|
270 | markers.append((chrom,abspos,snp)) # decorate for sort into genomic |
---|
271 | snpcols[snp] = snpIndex # so we know which col to find genos for this marker |
---|
272 | snpIndex += 1 |
---|
273 | markers.sort() |
---|
274 | rslist = [x[2] for x in markers] # drop decoration |
---|
275 | rsdict = dict(zip(rslist,rslist)) |
---|
276 | mfile.close() |
---|
277 | return markers,snpcols,rslist,rsdict |
---|
278 | |
---|