Mercurial > repos > davidvanzessen > argalaxy_tools

comparison report_clonality/RScript.r @ 6:d001d0c05dbe draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Fri, 16 Dec 2016 07:30:32 -0500
parents bcec7bb4e089
children 8cbc1a8d27ae
comparison
equal deleted inserted replaced
5:bcec7bb4e089 6:d001d0c05dbe
398 398
399 VandDCount$l = log(VandDCount$Length) 399 VandDCount$l = log(VandDCount$Length)
400 maxVD = data.frame(data.table(VandDCount)[, list(max=max(l)), by=c("Sample")]) 400 maxVD = data.frame(data.table(VandDCount)[, list(max=max(l)), by=c("Sample")])
401 VandDCount = merge(VandDCount, maxVD, by.x="Sample", by.y="Sample", all.x=T) 401 VandDCount = merge(VandDCount, maxVD, by.x="Sample", by.y="Sample", all.x=T)
402 VandDCount$relLength = VandDCount$l / VandDCount$max 402 VandDCount$relLength = VandDCount$l / VandDCount$max
403 check = is.nan(VandDCount$relLength)
404 if(any(check)){
405 VandDCount[check,"relLength"] = 0
406 }
403 407
404 cartegianProductVD = expand.grid(Top.V.Gene = Vchain$v.name, Top.D.Gene = Dchain$v.name) 408 cartegianProductVD = expand.grid(Top.V.Gene = Vchain$v.name, Top.D.Gene = Dchain$v.name)
405 409
406 completeVD = merge(VandDCount, cartegianProductVD, by.x=c("Top.V.Gene", "Top.D.Gene"), by.y=c("Top.V.Gene", "Top.D.Gene"), all=TRUE) 410 completeVD = merge(VandDCount, cartegianProductVD, by.x=c("Top.V.Gene", "Top.D.Gene"), by.y=c("Top.V.Gene", "Top.D.Gene"), all=TRUE)
407 411
443 447
444 VandJCount$l = log(VandJCount$Length) 448 VandJCount$l = log(VandJCount$Length)
445 maxVJ = data.frame(data.table(VandJCount)[, list(max=max(l)), by=c("Sample")]) 449 maxVJ = data.frame(data.table(VandJCount)[, list(max=max(l)), by=c("Sample")])
446 VandJCount = merge(VandJCount, maxVJ, by.x="Sample", by.y="Sample", all.x=T) 450 VandJCount = merge(VandJCount, maxVJ, by.x="Sample", by.y="Sample", all.x=T)
447 VandJCount$relLength = VandJCount$l / VandJCount$max 451 VandJCount$relLength = VandJCount$l / VandJCount$max
452
453 check = is.nan(VandJCount$relLength)
454 if(any(check)){
455 VandJCount[check,"relLength"] = 0
456 }
448 457
449 cartegianProductVJ = expand.grid(Top.V.Gene = Vchain$v.name, Top.J.Gene = Jchain$v.name) 458 cartegianProductVJ = expand.grid(Top.V.Gene = Vchain$v.name, Top.J.Gene = Jchain$v.name)
450 459
451 completeVJ = merge(VandJCount, cartegianProductVJ, all.y=TRUE) 460 completeVJ = merge(VandJCount, cartegianProductVJ, all.y=TRUE)
452 completeVJ = merge(completeVJ, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE) 461 completeVJ = merge(completeVJ, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
487 496
488 DandJCount$l = log(DandJCount$Length) 497 DandJCount$l = log(DandJCount$Length)
489 maxDJ = data.frame(data.table(DandJCount)[, list(max=max(l)), by=c("Sample")]) 498 maxDJ = data.frame(data.table(DandJCount)[, list(max=max(l)), by=c("Sample")])
490 DandJCount = merge(DandJCount, maxDJ, by.x="Sample", by.y="Sample", all.x=T) 499 DandJCount = merge(DandJCount, maxDJ, by.x="Sample", by.y="Sample", all.x=T)
491 DandJCount$relLength = DandJCount$l / DandJCount$max 500 DandJCount$relLength = DandJCount$l / DandJCount$max
501
502 check = is.nan(DandJCount$relLength)
503 if(any(check)){
504 DandJCount[check,"relLength"] = 0
505 }
492 506
493 cartegianProductDJ = expand.grid(Top.D.Gene = Dchain$v.name, Top.J.Gene = Jchain$v.name) 507 cartegianProductDJ = expand.grid(Top.D.Gene = Dchain$v.name, Top.J.Gene = Jchain$v.name)
494 508
495 completeDJ = merge(DandJCount, cartegianProductDJ, all.y=TRUE) 509 completeDJ = merge(DandJCount, cartegianProductDJ, all.y=TRUE)
496 completeDJ = merge(completeDJ, revDchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE) 510 completeDJ = merge(completeDJ, revDchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)