Supplemental Data 2 for Boot et al., “Characterization of colibactin-associated mutational signature in an Asian oral squamous cell carcinoma and in other mucosal tumor types”

This script applies the mSigAct signature presence test (Ng et al. 2017), to evaluate the hypothesis that the mutational signature identified in OSCC 62074759 (the A^nT signature) is present in Candidate tumors. Candidate tumors were previously identified as tumors showing strong enrichment for mutations with the characteristics of A^nT as described in Supplemental Data 1 of this manuscript.

Methods / code

We start with the following input data:

candidates is the tumors identified using the script in Supplementary Data 1 (this is the ‘results’ object from this script)
PCAWG attributions contain 4 mutation files, one for each of the 4 major groups of tumors in the PCAWG7 analysis of mutational signatures (Alexandrov et al., 2018, downloaded from: https://www.synapse.org/#!Synapse:syn11804065 on June 24th 2018).

First, extract for each of the candidates, the PCAWG7 signature assignments which will be later used in the analysis

library(stringr)
candidates<-read.csv("candidatesW_AnT.txt",sep="\t",as.is=T)
## remove 62074759
candidates<-candidates[-nrow(candidates),]
candidates$sample<-str_split_fixed(rownames(candidates),"___",2)[,1]

## rename the tumor types in the candidates table because of different annotations
tumorTypes<-read.csv("tumorTypes.txt",sep="\t",as.is=T,row.names = 1)
for(i in rownames(tumorTypes)){
  candidates$sample<-gsub(i,
                          tumorTypes$annotation[rownames(tumorTypes) == i],
                          candidates$sample)
}

## specify which file we should look in for the attributions
candidates$attributions[candidates$dataType == "WES_Other"]<-"nonPCAWG_WES"
candidates$attributions[candidates$dataType == "WES_TCGA"]<-"TCGA_WES"
candidates$attributions[candidates$dataType == "WGS_Other"]<-"nonPCAWG_WGS"
candidates$attributions[candidates$dataType == "WGS_ICGC"]<-"PCAWG_WGS"

## gather the signature attributions for the candidate samples
files<-list.files("PCAWG7 attributions",full.names = T)
attributions<-NULL
for(i in files){
  df<-read.csv(i,as.is=T)
  samples<-paste0(df$Cancer.Types,"::",df$Sample.Names)
  df2<-df[samples %in% candidates$sample,]
  attributions<-rbind(attributions,df2)
}

load scripts for mSigAct analysis and prep signatures and catalogs

source("src/mSigTools.v0.13.R")

## Loading required package: SnowballC

source("src/mSigAct.v0.10.R")

## 
## Attaching package: 'sets'

## The following object is masked from 'package:stringr':
## 
##     %>%

## load signatures
cosmic.sigs <- 
  get.signatures(
    signature.file='PCAWG7and62074759_96.tsv',
    exome.op=.h19.96.sureselect.v6.op)
cosmic.wes <- cosmic.sigs$exome
cosmic.wgs <- cosmic.sigs$genome
rm(cosmic.sigs)
# Downstream mSigAct requires that the elements of signatures sum to exactly 1. 
# Eventually move this code to mSigAct or change to all.equals tolerance=....
# We need to run multiple sweep call, presumably because of rounding.
for (i in 1:3) cosmic.wes <- sweep(cosmic.wes, MARGIN=2, colSums(cosmic.wes), '/')
stopifnot(colSums(cosmic.wes) == 1)
for (i in 1:3) cosmic.wgs <- sweep(cosmic.wgs, MARGIN=2, colSums(cosmic.wgs), '/')
stopifnot(colSums(cosmic.wgs) == 1)

## load the catalogs of the candidate tumors
exomes<-read.csv("catalogs/spectrum_counts_exomes.txt",sep="\t",as.is=T)
genomes<-read.csv("catalogs/spectrum_counts_genomes.txt",sep="\t",as.is=T)

Define function to run mSigAct per sample, using the signatures attributed in the PCAWG analysis and AnT.

sortSigs<-function(sigs,base="SBS"){
  sigs<-sigs[order(as.numeric(gsub("a","",
                                   gsub("b","",
                                        gsub("c","",
                                             gsub("d","",
                                                  gsub(base,"",sigs)))))))]
  return(sigs)
}

run.mSigAct.per.smp<-function(smp){
  tmp<-attributions[attributions$Sample.Names == smp,4:ncol(attributions)]
  sigs<-colnames(tmp)[!tmp[1,] == 0]
  
  ## signatures SBS1, SBS5 and SBS40 are present in all tumors 
  ## therefore regardless of previous assignments, 
  ## add SBS1 and SBS5 in this analysis
  if(!"SBS1" %in% sigs){sigs<-c(sigs,"SBS1")}
  if(!"SBS5" %in% sigs){sigs<-c(sigs,"SBS5")}
  sigs<-sortSigs(sigs)

  ## depending on data type, use WGS or WES signatures
  dataType<-candidates$data[grep(smp,candidates$sample)]
  dataType<-str_split_fixed(dataType,"_",2)[,1]
  
  if(dataType == "WGS"){ 
    universe<-cosmic.wgs
    catalog<-genomes
  } else {
      universe<-cosmic.wes
      catalog<-exomes
  }
  subverse<-universe[,c(sigs,"AnT")]
  
  ## ensure that catalog and signatures are in the same order
  rownames(catalog)<-paste0(catalog$Before,catalog$Ref,
                            catalog$After,catalog$Var)
  subverse<-subverse[rownames(catalog),]
  input<-as.matrix(catalog[,grep(gsub("-",".",smp),colnames(catalog))])
  colnames(input)<-smp
  rownames(input)<-rownames(subverse)
  
  analysis<-process.one.group(input, 
                             subverse,
                             target.sig.name = "AnT",
                             path.root=paste0("mSigAct_output/",smp),
                             obj.fun = obj.fun.nbinom.maxlh,
                             nbinom.size=10, ## = dispersion parameter
                             mc.cores=1)     ## = number of cores 
  
  # results
  df<-t(rbind(pval=analysis$pval,analysis$exposure))
  rownames(df)<-smp
  return(df)
}

Run mSigAct per sample and generate a summary results table

## make a df to save the attributions in
mSigAct_result<-attributions
mSigAct_result$AnT<-NA
mSigAct_result$mSigAct_pval<-0

for(smp in attributions$Sample.Names){
  df<-as.data.frame(run.mSigAct.per.smp(smp))
  mSigAct_result$mSigAct_pval[mSigAct_result$Sample.Names == smp]<-
    df$pval
  mSigAct_result[mSigAct_result$Sample.Names == smp,
                 colnames(mSigAct_result) %in% colnames(df)]<-
    df[1,2:ncol(df)]
}

## remove signatures that are not attributed to any sample
mSigAct_result<-mSigAct_result[c(2,which(colSums(mSigAct_result[,-c(1:3)])>0)+3)]
## perform multiple testing correction
mSigAct_result$mSigAct_qval<-p.adjust(mSigAct_result$mSigAct_pval,method="BH")

## format to print output table to html
for(i in 2:(ncol(mSigAct_result)-2)){mSigAct_result[,i]<-as.numeric(format(as.numeric(mSigAct_result[,i]),digits=0,scientific=FALSE))}
mSigAct_result$mSigAct_pval<-format(mSigAct_result$mSigAct_pval,digits=3)
mSigAct_result$mSigAct_qval<-format(mSigAct_result$mSigAct_qval,digits=3)
library(knitr)
kable(mSigAct_result[,-(ncol(mSigAct_result)-1)],row.names=F,align = c("l",rep("c",14)))

Sample.Names	SBS1	SBS2	SBS5	SBS7a	SBS7b	SBS9	SBS10a	SBS10b	SBS13	SBS14	SBS15	SBS16	SBS17a	SBS17b	SBS18	SBS27	SBS28	SBS37	SBS40	SBS45	SBS57	AnT	mSigAct_qval
BD121T	117	0	145	0	0	0	1119	833	0	0	0	0	0	0	0	0	0	0	0	0	0	70	3.27e-03
BD173T	0	0	0	0	0	0	0	0	0	0	0	377	0	0	0	0	0	0	0	0	0	846	6.48e-39
BD182T	0	0	349	0	0	0	0	0	0	0	0	0	0	0	0	0	879	0	0	0	0	877	1.77e-69
BD223T	0	0	160	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	80	182	2.40e-16
sysucc-311T	0	0	745	0	0	0	3913	7406	0	0	0	0	0	0	0	0	1864	0	0	0	0	231	8.01e-04
ESO-173	64	0	82	0	0	0	0	0	0	0	0	0	23	33	0	0	0	0	0	0	0	0	1.00e+00
HCC34T	0	0	249	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1.75e-01
PCSI_0060_Pa_X	33	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	32	47	6.68e-07
SKCM-JWCI-WGS-8-Tumor	0	0	125	394	187	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	9.27e-01
T155	37	0	41	0	0	0	467	438	0	0	0	0	0	0	0	0	0	0	0	0	0	0	3.25e-01
LP6005935-DNA_B03	899	0	11769	0	0	0	0	0	0	0	0	0	5759	16897	0	0	0	0	0	0	0	9577	3.63e-18
8069334	1153	0	3229	0	0	0	0	0	0	0	0	0	500	1357	0	0	0	0	0	0	0	862	3.01e-07
0047_CRUK_PC_0047_T1_DNA	496	0	1904	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	951	1.85e-16
SP22031	2490	0	4530	0	0	0	124275	80170	0	0	0	0	0	0	0	0	22766	0	0	0	0	0	4.54e-01
SP16886	0	0	12505	0	0	0	434277	251355	0	0	0	0	0	0	0	0	151145	0	0	0	0	0	7.13e-01
SP19295	4183	0	6759	0	0	0	127419	102190	0	0	0	0	0	0	0	0	17351	0	0	0	0	1965	7.91e-02
SP17905	0	0	30073	0	0	0	1215032	478155	0	0	0	0	0	0	0	0	523625	185733	0	0	0	0	8.33e-01
SP21400	9932	0	15411	0	0	0	256130	446142	0	0	0	0	0	0	0	0	65424	0	0	0	0	0	1.75e-01
SP18946	0	0	16982	0	0	0	490500	366755	0	0	0	0	0	0	0	0	86576	0	0	0	0	0	4.54e-01
SP80615	0	0	57839	0	0	0	1070130	795814	0	0	0	0	0	0	0	0	479938	0	0	0	0	22748	2.38e-02
SP81494	2341	0	12182	0	0	0	0	0	0	0	0	0	0	0	2613	0	0	0	2963	4521	0	2148	7.18e-04
SP81711	3789	0	4065	0	0	0	0	0	0	0	0	0	0	0	3537	0	0	0	4975	0	0	3908	1.08e-12
SP80754	2305	0	2948	0	0	0	0	0	0	0	0	0	0	0	2097	0	206	0	0	0	0	3653	6.48e-39
SP111026	1481	0	8490	0	0	0	0	0	0	0	0	0	9010	24066	0	0	0	0	0	0	0	3742	2.77e-10
SP111101	3450	0	3123	0	0	0	0	0	0	0	0	0	1011	1607	0	0	0	0	5476	0	0	2245	4.04e-07
SP92659	1802	0	6991	0	0	0	191629	55154	0	0	0	0	0	0	0	0	26686	0	0	0	0	0	4.54e-01
TCGA-AB-2824-03B-01W-0728-08	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	70	0	0	0	0	0	0	1.00e+00
TCGA-AB-2851-03B-01W-0728-08	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	58	0	0	0	0	0	0	1.00e+00
TCGA-AB-2867-03B-01W-0728-08	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	115	0	0	0	0	0	0	1.00e+00
TCGA-AB-2868-03B-01W-0728-08	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	201	0	0	0	0	0	3	9.49e-02
TCGA-FU-A3HZ-01A-11D-A20U-09	0	0	107	0	0	0	1418	1003	0	0	0	0	0	0	0	0	655	0	0	0	0	71	1.61e-03
TCGA-AY-4071-01A-01W-1073-09	45	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	79	0	0	40	7.69e-05
TCGA-AG-3892-01A-01W-1073-09	0	0	91	0	0	0	1932	1570	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1.00e+00
TCGA-AG-3902-01A-01W-1073-09	42	0	41	0	0	20	0	0	8	0	0	0	0	0	0	0	0	0	0	0	0	65	1.58e-07
TCGA-2H-A9GM-01A-11D-A37C-09	23	0	81	0	0	0	0	0	0	0	0	0	5	21	43	0	0	0	0	0	0	0	2.38e-01
TCGA-IG-A4QS-01A-11D-A27G-09	69	0	112	0	0	0	0	0	0	0	0	0	38	64	0	0	0	0	0	0	0	36	2.08e-02
TCGA-R6-A6L4-01A-11D-A31U-09	8	0	63	0	0	0	0	0	10	0	0	0	17	44	19	0	0	0	0	0	0	0	7.13e-01
TCGA-BA-A4IG-01A-11D-A25Y-08	0	16	81	0	0	0	0	0	27	0	0	0	0	0	0	0	0	0	0	0	0	44	1.51e-06
TCGA-BR-6453-01A-11D-1800-08	49	0	81	0	0	0	0	0	0	0	0	0	32	65	0	0	0	0	0	0	0	0	1.00e+00
TCGA-D7-A4Z0-01A-22D-A25D-08	37	0	62	0	0	0	0	0	0	0	0	0	97	169	0	0	0	0	0	0	0	0	6.07e-01
TCGA-G2-AA3B-01A-11D-A391-08	0	352	101	0	0	0	0	0	438	0	0	0	0	0	0	0	0	0	0	0	0	45	3.25e-04
TCGA-GC-A6I3-01A-11D-A31L-08	12	71	105	0	0	0	0	0	90	0	0	0	0	0	0	0	0	0	0	0	0	32	3.89e-03
TCGA-A5-A0GP-01A-11W-A062-09	76	0	202	0	0	0	950	804	0	0	0	0	0	0	0	0	195	0	0	0	0	0	3.25e-01
TCGA-AJ-A5DW-01A-11D-A27P-09	97	0	307	0	0	0	3352	1499	0	0	0	0	0	0	0	0	366	0	0	0	0	0	4.15e-01
TCGA-AP-A1E0-01A-11D-A135-09	216	0	325	0	0	0	4978	2070	0	0	0	0	0	0	0	0	810	0	0	0	0	0	9.48e-01
TCGA-AX-A1CE-01A-11D-A135-09	0	0	190	0	0	0	0	0	0	4221	15172	0	0	0	0	0	0	0	0	0	0	0	1.13e-01
TCGA-BK-A6W3-01A-12D-A34Q-09	123	0	335	0	0	0	4032	1867	0	0	0	0	0	0	0	0	722	0	0	0	0	90	4.80e-02
TCGA-DF-A2KV-01A-11D-A17W-09	38	0	74	0	0	0	1008	938	0	0	0	0	0	0	0	0	286	0	0	0	0	47	2.34e-02
TCGA-E6-A1M0-01A-11D-A142-09	56	0	126	0	0	0	1913	778	0	0	0	0	0	0	0	0	296	0	0	0	0	0	4.55e-01
TCGA-EO-A3AV-01A-12D-A19Y-09	0	0	398	0	0	0	4037	1899	0	0	0	0	0	0	0	0	990	0	0	0	0	0	1.94e-01
TCGA-EO-A3AY-01A-12D-A19Y-09	0	0	333	0	0	0	3805	1489	0	0	0	0	0	0	0	0	686	0	0	0	0	0	1.13e-01
TCGA-EY-A1GD-01A-11D-A13L-09	57	0	173	0	0	0	581	771	0	0	0	0	0	0	0	0	274	0	0	0	0	38	7.91e-02
TCGA-EY-A1GI-01A-11D-A13L-09	101	0	421	0	0	0	4053	2481	0	0	0	0	0	0	0	0	818	0	0	0	0	0	2.07e-01
TCGA-QF-A5YS-01A-11D-A31U-09	115	0	110	0	0	0	1152	1141	0	0	0	0	0	0	0	0	220	0	0	0	0	50	2.38e-02
TCGA-QS-A5YQ-01A-11D-A31U-09	58	0	128	0	0	0	857	485	0	0	0	0	0	0	0	0	142	0	0	0	0	0	7.23e-01