Bhattacharya authored on 20/08/2021 20:18:52 | 1 | 1 |
new file mode 100644 |
| ... | ... |
@@ -0,0 +1,1365 @@ |
| 1 |
+#' Combining the RNAseq reads of family members in a |
|
| 2 |
+#' single file. |
|
| 3 |
+#' |
|
| 4 |
+#' @param RNASeqDir character. Directory containing RNAseq reads. |
|
| 5 |
+#' @param returnMethod character. Method of returning Data. |
|
| 6 |
+#' @param outpath character. Contains file path if Method of return is chosen as |
|
| 7 |
+#' Text. |
|
| 8 |
+#' @param outFileName character. Output file name. |
|
| 9 |
+#' @return Text or Dataframe containing TPM read counts of genes in the family. |
|
| 10 |
+#' @examples |
|
| 11 |
+#' \dontrun{
|
|
| 12 |
+#' RNASeqDir = system.file("extdata", package="nanotatoR")
|
|
| 13 |
+#' returnMethod="dataFrame" |
|
| 14 |
+#' datRNASeq <- RNAseqcombine(RNASeqDir = RNASeqDir, |
|
| 15 |
+#' returnMethod = returnMethod) |
|
| 16 |
+#' } |
|
| 17 |
+#' @importFrom stats na.omit |
|
| 18 |
+#' @importFrom AnnotationDbi mapIds |
|
| 19 |
+#' @import org.Hs.eg.db |
|
| 20 |
+#' @export |
|
| 21 |
+RNAseqcombine<-function(RNASeqDir,returnMethod=c("Text","dataFrame"),
|
|
| 22 |
+ outpath="",outFileName=""){
|
|
| 23 |
+ #library(biomaRt) |
|
| 24 |
+ #setwd(RNASeqDir) |
|
| 25 |
+ l <- list.files(path = RNASeqDir, |
|
| 26 |
+ pattern="*.genes.results", full.names = FALSE) |
|
| 27 |
+ len<-length(l) |
|
| 28 |
+ #dat<-listDatasets(ensembl) |
|
| 29 |
+ #g1<-grep("sscrofa",listDatasets(ensembl)$dataset)
|
|
| 30 |
+ 'grch37 = useMart(biomart="ENSEMBL_MART_ENSEMBL", host="www.ensembl.org", |
|
| 31 |
+ dataset="hsapiens_gene_ensembl")' |
|
| 32 |
+ gen<-c(); |
|
| 33 |
+ ##Need to make this function dynamic |
|
| 34 |
+ dat<-data.frame(matrix(ncol = len, |
|
| 35 |
+ nrow = nrow(r<-read.table(file.path(RNASeqDir,l[1]),sep="\t",header=TRUE)))) |
|
| 36 |
+ cnam<-c() |
|
| 37 |
+ for (ii in 1:length(l)){
|
|
| 38 |
+ r <- read.table(file.path(RNASeqDir,l[ii]),sep="\t",header=TRUE) |
|
| 39 |
+ gen <- c(gen,as.character(r$gene_id)) |
|
| 40 |
+ dat[,ii] <- as.numeric(r$TPM) |
|
| 41 |
+ str <- strsplit(l[ii],split=".genes.results") |
|
| 42 |
+ #print(str[[1]][1]) |
|
| 43 |
+ cnam <- c(cnam,str[[1]][1]) |
|
| 44 |
+ } |
|
| 45 |
+ #datf<-data.frame(dat) |
|
| 46 |
+ gen <- unique(as.character(gen)) |
|
| 47 |
+ st <- strsplit(gen,split="[.]") |
|
| 48 |
+ genes <- c() |
|
| 49 |
+ for(k in 1:length(gen)){
|
|
| 50 |
+ genes<-c(genes,as.character(st[[k]][1])) |
|
| 51 |
+ } |
|
| 52 |
+ data1<-dat[,1:ncol(dat)] |
|
| 53 |
+ names(data1)<-cnam |
|
| 54 |
+ genesym<-c() |
|
| 55 |
+ ensemblid<-c() |
|
| 56 |
+ 'gene1 = getBM(attributes = c("external_gene_name", "ensembl_gene_id"),
|
|
| 57 |
+ filters = "ensembl_gene_id", values = genes, mart = grch37)' |
|
| 58 |
+ gn1 <- mapIds(org.Hs.eg.db, genes, "SYMBOL", "ENSEMBL") |
|
| 59 |
+ 'rn <- row.names(data.frame(gn1)) |
|
| 60 |
+ rn1 <- row.names(data.frame(gn2))' |
|
| 61 |
+ gene1<-data.frame( |
|
| 62 |
+ ensembl_gene_id = as.character(names((gn1))), |
|
| 63 |
+ external_gene_name = as.character(data.frame(gn1)[,1]) |
|
| 64 |
+ ) |
|
| 65 |
+ genesym<-as.character(gene1$external_gene_name) |
|
| 66 |
+ ensemblid<-as.character(gene1$ensembl_gene_id) |
|
| 67 |
+ gene3<-c() |
|
| 68 |
+ ens<-c() |
|
| 69 |
+ ensemblid<-paste("^",ensemblid,"$",sep="")
|
|
| 70 |
+ for(kk in 1:length(genes)){
|
|
| 71 |
+ pag<-paste("^",genes[kk],"$",sep="")
|
|
| 72 |
+ val<-grep(pag,ensemblid,fixed=TRUE) |
|
| 73 |
+ #if(length(val) > 1) {print(paste("val:", val, "kk:", kk))}
|
|
| 74 |
+ if(length(val)>0){
|
|
| 75 |
+ gene3<-c(gene3,as.character(unique(genesym[val]))) |
|
| 76 |
+ ens<-c(ens,as.character(rep(genes[kk], length(unique(genesym[val])) |
|
| 77 |
+ ))) |
|
| 78 |
+ } |
|
| 79 |
+ else{
|
|
| 80 |
+ gene3<-c(gene3,as.character("-"))
|
|
| 81 |
+ ens<-c(ens,as.character(genes[kk])) |
|
| 82 |
+ } |
|
| 83 |
+ } |
|
| 84 |
+ RNASeqDat<-data.frame(GeneName = as.character(gene3), |
|
| 85 |
+ GeneID = as.character(ens),data1) |
|
| 86 |
+ if(returnMethod == "Text"){
|
|
| 87 |
+ fname = file.path(outFileName, ".csv" ,sep = "") |
|
| 88 |
+ write.csv(RNASeqDat,file.path(outpath,fname), |
|
| 89 |
+ row.names = FALSE) |
|
| 90 |
+ } |
|
| 91 |
+ else if (returnMethod == "dataFrame"){
|
|
| 92 |
+ return (RNASeqDat) |
|
| 93 |
+ } |
|
| 94 |
+ else{
|
|
| 95 |
+ stop("Invalid ReturnMethod")
|
|
| 96 |
+ } |
|
| 97 |
+} |
|
| 98 |
+ |
|
| 99 |
+#' Extract Read counts for genes that overlap SVs. |
|
| 100 |
+#' |
|
| 101 |
+#' @param gnsOverlap character. genes that overlap SV. |
|
| 102 |
+#' @param SVID character. ID of the SVs. |
|
| 103 |
+#' @param RNASeqData character. Expression of the genes. |
|
| 104 |
+#' @param pattern_Proband character. Pattern to identify the proband reads. |
|
| 105 |
+#' @param pattern_Father character. Pattern to identify the father reads. |
|
| 106 |
+#' @param pattern_Mother character. Pattern to identify the mother reads. |
|
| 107 |
+#' @return Text or Dataframe containing TPM read counts of genes in the family. |
|
| 108 |
+#' @examples |
|
| 109 |
+#' RNASeqDir = system.file("extdata", package="nanotatoR")
|
|
| 110 |
+#' returnMethod="dataFrame" |
|
| 111 |
+#' datRNASeq <- RNAseqcombine(RNASeqDir = RNASeqDir, |
|
| 112 |
+#' returnMethod = returnMethod) |
|
| 113 |
+#' gnsOverlap <- c("AGL")
|
|
| 114 |
+#' SVID = 397 |
|
| 115 |
+#' datgnovrlap <- OverlapRNAseq(gnsOverlap = gnsOverlap, |
|
| 116 |
+#' SVID = SVID, RNASeqData = datRNASeq, |
|
| 117 |
+#' pattern_Proband = "*_P_*") |
|
| 118 |
+#' @importFrom stats na.omit |
|
| 119 |
+#' @export |
|
| 120 |
+OverlapRNAseq<-function(gnsOverlap, SVID, RNASeqData, |
|
| 121 |
+ pattern_Proband = NA, pattern_Mother = NA, |
|
| 122 |
+ pattern_Father = NA){
|
|
| 123 |
+ |
|
| 124 |
+ ###Finding the column names |
|
| 125 |
+ #print(gnsOverlap);print(length(gnsOverlap)) |
|
| 126 |
+ if(is.na(unique(as.character(pattern_Father))) == FALSE){
|
|
| 127 |
+ fatherInd<-grep(unique(as.character(pattern_Father)),names(RNASeqData)) |
|
| 128 |
+ } else{fatherInd<- NA}
|
|
| 129 |
+ if(is.na(unique(as.character(pattern_Mother))) == FALSE){
|
|
| 130 |
+ motherInd<-grep(unique(as.character(pattern_Mother)),names(RNASeqData)) |
|
| 131 |
+ } else{motherInd<- NA}
|
|
| 132 |
+ if(is.na(unique(as.character(pattern_Proband))) == FALSE){
|
|
| 133 |
+ probandInd<-grep(unique(as.character(pattern_Proband)),names(RNASeqData)) |
|
| 134 |
+ } else{probandInd<- NA}
|
|
| 135 |
+ 'if(is.na(pattern_Sibling)==FALSE){
|
|
| 136 |
+ siblingInd<-grep(pattern_Sibling,names(RNASeqData)) |
|
| 137 |
+ } |
|
| 138 |
+ else{
|
|
| 139 |
+ siblingInd<-NA |
|
| 140 |
+ }' |
|
| 141 |
+ sv<-c() |
|
| 142 |
+ gene<-c() |
|
| 143 |
+ gnsname<-as.character(RNASeqData$GeneName) |
|
| 144 |
+ pasgnsname<-pasgnovlap<-paste("^",gnsname,"$",sep="")
|
|
| 145 |
+ 'overlap_ensemblgenes = select(EnsDb.Hsapiens.v79, gnsOverlap, |
|
| 146 |
+ c("GENEID","GENENAME"), "SYMBOL")
|
|
| 147 |
+ gnsOverlapID<-as.character(overlap_ensemblgenes$GENEID)' |
|
| 148 |
+ #print(paste("gnsOverlap:",gnsOverlap))
|
|
| 149 |
+ #print(paste("overlap_ensemblgenes:",overlap_ensemblgenes))
|
|
| 150 |
+ #print(paste("gnsOverlapID:",gnsOverlapID))
|
|
| 151 |
+ #genes<-as.character(overlap_ensemblgenes$SYMBOL) |
|
| 152 |
+ ###Extracting Reads |
|
| 153 |
+ ### |
|
| 154 |
+ ###Genes Names Extraction |
|
| 155 |
+ #print(paste("fatherInd :",fatherInd))
|
|
| 156 |
+ #print(paste("motherInd :",motherInd))
|
|
| 157 |
+ #print(paste("probandInd :",probandInd))
|
|
| 158 |
+ #print(paste("siblingInd :",siblingInd))
|
|
| 159 |
+ gnsOverlapID <- as.character(gnsOverlap) |
|
| 160 |
+ #print(gnsOverlapID) |
|
| 161 |
+ if(length(gnsOverlapID)>1){
|
|
| 162 |
+ |
|
| 163 |
+ datGeneInfoTemp<-data.frame() |
|
| 164 |
+ fatherReads<-c() |
|
| 165 |
+ motherReads<-c() |
|
| 166 |
+ probandReads<-c() |
|
| 167 |
+ #siblingReads<-c() |
|
| 168 |
+ for (ki in 1:length(gnsOverlapID)){
|
|
| 169 |
+ pasgnovlap<-paste("^",gnsOverlapID[ki],"$",sep="")
|
|
| 170 |
+ #print(ki) |
|
| 171 |
+ gg<-grep(pasgnovlap,pasgnsname,fixed=TRUE) |
|
| 172 |
+ dat_temp<-RNASeqData[gg,] |
|
| 173 |
+ |
|
| 174 |
+ if(nrow(dat_temp)>1){
|
|
| 175 |
+ dat_temp_1<-apply(dat_temp[,3:ncol(dat_temp)],2,mean) |
|
| 176 |
+ #dat_temp_1<-data.frame(dat_temp_1) |
|
| 177 |
+ dat_temp1<-dat_temp[1,] |
|
| 178 |
+ dat_temp1[,3:ncol(dat_temp)]<-dat_temp_1 |
|
| 179 |
+ #print(dim(dat_temp1)) |
|
| 180 |
+ fathercount<-c();mothercount<-c(); |
|
| 181 |
+ probandcount<-c(); |
|
| 182 |
+ if(is.na(fatherInd[1])== FALSE){
|
|
| 183 |
+ if(length(fatherInd)>1){
|
|
| 184 |
+ for(j in fatherInd){
|
|
| 185 |
+ fathercount<-c(fathercount,dat_temp1[,j]) |
|
| 186 |
+ } |
|
| 187 |
+ fatherReads<-c(fatherReads, |
|
| 188 |
+ paste(fathercount,collapse = ":")) |
|
| 189 |
+ }else if(length(fatherInd)==1){
|
|
| 190 |
+ fatherReads<-c(fatherReads,dat_temp1[,fatherInd]) |
|
| 191 |
+ }else{
|
|
| 192 |
+ fatherReads<-c(fatherReads,0) |
|
| 193 |
+ } |
|
| 194 |
+ }else{
|
|
| 195 |
+ fatherReads<-c(fatherReads,"-") |
|
| 196 |
+ } |
|
| 197 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 198 |
+ if(length(motherInd)>1){
|
|
| 199 |
+ for(j in motherInd){
|
|
| 200 |
+ mothercount<-c(mothercount,dat_temp1[,j]) |
|
| 201 |
+ } |
|
| 202 |
+ motherReads<-c(motherReads,paste(mothercount, |
|
| 203 |
+ collapse = ":")) |
|
| 204 |
+ }else if(length(motherInd)==1){
|
|
| 205 |
+ motherReads<-c(motherReads,dat_temp1[,motherInd]) |
|
| 206 |
+ }else{
|
|
| 207 |
+ motherReads<-c(motherReads,0) |
|
| 208 |
+ } |
|
| 209 |
+ }else{
|
|
| 210 |
+ motherReads<-c(motherReads,"-") |
|
| 211 |
+ } |
|
| 212 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 213 |
+ if(length(probandInd)>1){
|
|
| 214 |
+ for(j in probandInd){
|
|
| 215 |
+ probandcount<-c(probandcount,dat_temp1[,j]) |
|
| 216 |
+ } |
|
| 217 |
+ probandReads<-c(probandReads,paste(probandcount,collapse=":")) |
|
| 218 |
+ }else if(length(probandInd)==1){
|
|
| 219 |
+ probandReads<-c(probandReads,dat_temp1[,probandInd]) |
|
| 220 |
+ }else{
|
|
| 221 |
+ probandReads<-c(probandReads,0) |
|
| 222 |
+ } |
|
| 223 |
+ }else{
|
|
| 224 |
+ probandReads<-c(probandReads, "-") |
|
| 225 |
+ } |
|
| 226 |
+ |
|
| 227 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 228 |
+ if(length(siblingInd)>1){
|
|
| 229 |
+ for(j in siblingInd){
|
|
| 230 |
+ siblingcount<-c(siblingcount,dat_temp1[,j]) |
|
| 231 |
+ } |
|
| 232 |
+ siblingReads<-c(siblingReads,paste(siblingcount,collapse=":")) |
|
| 233 |
+ } |
|
| 234 |
+ else if(length(siblingInd)==1){
|
|
| 235 |
+ siblingReads<-c(siblingReads,dat_temp1[,siblingInd]) |
|
| 236 |
+ } |
|
| 237 |
+ else{
|
|
| 238 |
+ siblingReads<-c(siblingReads,0) |
|
| 239 |
+ } |
|
| 240 |
+ } |
|
| 241 |
+ else{
|
|
| 242 |
+ siblingReads<-c(siblingReads,"-") |
|
| 243 |
+ }' |
|
| 244 |
+ } |
|
| 245 |
+ else if (nrow(dat_temp)==1) {
|
|
| 246 |
+ #print(dim(dat_temp1)) |
|
| 247 |
+ fathercount<-c();mothercount<-c();probandcount<-c(); |
|
| 248 |
+ if(is.na(fatherInd)==FALSE){
|
|
| 249 |
+ if(length(fatherInd [1])>1){
|
|
| 250 |
+ for(j in fatherInd){
|
|
| 251 |
+ fathercount<-c(fathercount,mean(dat_temp[,j])) |
|
| 252 |
+ } |
|
| 253 |
+ fatherReads<-c(fatherReads,paste(fathercount,collapse=":")) |
|
| 254 |
+ } |
|
| 255 |
+ else if(length(fatherInd)==1){
|
|
| 256 |
+ fatherReads<-c(fatherReads,mean(dat_temp[,fatherInd])) |
|
| 257 |
+ } |
|
| 258 |
+ else{
|
|
| 259 |
+ fatherReads<-c(fatherReads,0) |
|
| 260 |
+ } |
|
| 261 |
+ } |
|
| 262 |
+ else{
|
|
| 263 |
+ fatherReads<-c(fatherReads,"-") |
|
| 264 |
+ } |
|
| 265 |
+ if(is.na(motherInd [1])==FALSE){
|
|
| 266 |
+ if(length(motherInd)>1){
|
|
| 267 |
+ for(j in motherInd){
|
|
| 268 |
+ mothercount<-c(mothercount,dat_temp[,j]) |
|
| 269 |
+ } |
|
| 270 |
+ motherReads<-c(motherReads,paste(mothercount,collapse=":")) |
|
| 271 |
+ } |
|
| 272 |
+ else if(length(motherInd)==1){
|
|
| 273 |
+ motherReads<-c(motherReads,mean(dat_temp[,motherInd])) |
|
| 274 |
+ } |
|
| 275 |
+ else{
|
|
| 276 |
+ motherReads<-c(motherReads,0) |
|
| 277 |
+ } |
|
| 278 |
+ } |
|
| 279 |
+ else{
|
|
| 280 |
+ motherReads<-c(motherReads, "-") |
|
| 281 |
+ } |
|
| 282 |
+ if(is.na(probandInd [1])==FALSE){
|
|
| 283 |
+ if(length(probandInd)>1){
|
|
| 284 |
+ for(j in probandInd){
|
|
| 285 |
+ probandcount<-c(probandcount,dat_temp[,j]) |
|
| 286 |
+ } |
|
| 287 |
+ probandReads<-c(probandReads,paste(probandcount,collapse=":")) |
|
| 288 |
+ } |
|
| 289 |
+ else if(length(probandInd)==1){
|
|
| 290 |
+ probandReads<-c(probandReads,mean(dat_temp[,probandInd])) |
|
| 291 |
+ } |
|
| 292 |
+ else{
|
|
| 293 |
+ probandReads<-c(probandReads,0) |
|
| 294 |
+ } |
|
| 295 |
+ } |
|
| 296 |
+ else{
|
|
| 297 |
+ probandReads<-c(probandReads, "-") |
|
| 298 |
+ } |
|
| 299 |
+ |
|
| 300 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 301 |
+ if(length(siblingInd)>1){
|
|
| 302 |
+ for(j in siblingInd){
|
|
| 303 |
+ siblingcount<-c(siblingcount,dat_temp[,j]) |
|
| 304 |
+ } |
|
| 305 |
+ siblingReads<-c(siblingReads,paste(siblingcount,collapse=":")) |
|
| 306 |
+ } |
|
| 307 |
+ else if(length(siblingInd)==1){
|
|
| 308 |
+ siblingReads<-c(siblingReads,dat_temp[,siblingInd]) |
|
| 309 |
+ } |
|
| 310 |
+ else{
|
|
| 311 |
+ siblingReads<-c(siblingReads,0) |
|
| 312 |
+ } |
|
| 313 |
+ } |
|
| 314 |
+ else{
|
|
| 315 |
+ siblingReads<-c(siblingReads,"-") |
|
| 316 |
+ }' |
|
| 317 |
+ } |
|
| 318 |
+ else{
|
|
| 319 |
+ fatherReads<-c(fatherReads,"-") |
|
| 320 |
+ motherReads<-c(motherReads,"-") |
|
| 321 |
+ probandReads<-c(probandReads,"-") |
|
| 322 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 323 |
+ siblingReads<-c(siblingReads,"-") |
|
| 324 |
+ } |
|
| 325 |
+ else{
|
|
| 326 |
+ siblingReads<-c(siblingReads,"-") |
|
| 327 |
+ }' |
|
| 328 |
+ } |
|
| 329 |
+ gene<-c(gene,as.character(gnsOverlapID[ki])) |
|
| 330 |
+ } |
|
| 331 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 332 |
+ ProbandGenes<-c() |
|
| 333 |
+ for(ii in 1:length(gene)){
|
|
| 334 |
+ pasgene<-paste(gene[ii],"(",probandReads[ii],")",sep="")
|
|
| 335 |
+ ProbandGenes<-c(ProbandGenes,pasgene) |
|
| 336 |
+ } |
|
| 337 |
+ ProbandTPM<-paste(ProbandGenes,collapse=";") |
|
| 338 |
+ } else{
|
|
| 339 |
+ ProbandTPM <- "-" |
|
| 340 |
+ } |
|
| 341 |
+ |
|
| 342 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 343 |
+ FatherGenes<-c() |
|
| 344 |
+ for(ii in 1:length(gene)){
|
|
| 345 |
+ pasgene<-paste(gene[ii],"(",fatherReads[ii],")",sep="")
|
|
| 346 |
+ FatherGenes<-c(FatherGenes,pasgene) |
|
| 347 |
+ } |
|
| 348 |
+ FatherTPM<-paste(FatherGenes,collapse=";") |
|
| 349 |
+ } else{
|
|
| 350 |
+ FatherTPM <- "-" |
|
| 351 |
+ } |
|
| 352 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 353 |
+ MotherGenes<-c() |
|
| 354 |
+ for(ii in 1:length(gene)){
|
|
| 355 |
+ pasgene<-paste(gene[ii],"(",motherReads[ii],")",sep="")
|
|
| 356 |
+ MotherGenes<-c(MotherGenes,pasgene) |
|
| 357 |
+ } |
|
| 358 |
+ MotherTPM<-paste(MotherGenes,collapse=";") |
|
| 359 |
+ } else{
|
|
| 360 |
+ MotherTPM <- "-" |
|
| 361 |
+ } |
|
| 362 |
+ |
|
| 363 |
+ |
|
| 364 |
+ ' if(is.na(siblingInd[1])==FALSE){
|
|
| 365 |
+ siblingGenes<-c() |
|
| 366 |
+ for(ii in 1:length(gene)){
|
|
| 367 |
+ |
|
| 368 |
+ pasgene<-paste(gene[ii],"(",siblingReads[ii],")",sep="")
|
|
| 369 |
+ siblingGenes<-c(siblingGenes,pasgene) |
|
| 370 |
+ } |
|
| 371 |
+ SiblingTPM<-paste(siblingGenes,collapse=";") |
|
| 372 |
+ } |
|
| 373 |
+ else{
|
|
| 374 |
+ SiblingTPM<-"-" |
|
| 375 |
+ } |
|
| 376 |
+ ' |
|
| 377 |
+ datGeneInfo<-data.frame(SVID=SVID,Probandexpression=ProbandTPM, |
|
| 378 |
+ Fatherexpression=FatherTPM,Motherexpression=MotherTPM) |
|
| 379 |
+ |
|
| 380 |
+ } |
|
| 381 |
+ else if(length(gnsOverlapID)==1){
|
|
| 382 |
+ pasgnovlap<-paste("^",as.character(gnsOverlapID),"$",sep="")
|
|
| 383 |
+ #print(ki) |
|
| 384 |
+ gg<-grep(pasgnovlap,pasgnsname,fixed=TRUE) |
|
| 385 |
+ #gg<-grep(gnsOverlapID,gnsname,fixed=TRUE) |
|
| 386 |
+ dat_temp<-RNASeqData[gg,] |
|
| 387 |
+ if(nrow(dat_temp)>1){
|
|
| 388 |
+ dat_temp_1<-apply(dat_temp[,3:ncol(dat_temp)],2,mean) |
|
| 389 |
+ #dat_temp_1<-data.frame(dat_temp_1) |
|
| 390 |
+ dat_temp1<-dat_temp[1,] |
|
| 391 |
+ dat_temp1[,3:ncol(dat_temp)]<-dat_temp_1 |
|
| 392 |
+ #dat_temp1<-cbind(dat_temp[1,1:3],dat_temp1) |
|
| 393 |
+ #print(dim(dat_temp1)) |
|
| 394 |
+ fathercount<-c();mothercount<-c();probandcount<-c(); |
|
| 395 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 396 |
+ if(length(fatherInd)>=1){
|
|
| 397 |
+ for(j in fatherInd){
|
|
| 398 |
+ fathercount<-c(fathercount,mean(dat_temp1[,j])) |
|
| 399 |
+ } |
|
| 400 |
+ fatherReads<-paste(fathercount,collapse=":") |
|
| 401 |
+ } |
|
| 402 |
+ else if(length(fatherInd)==1){
|
|
| 403 |
+ fatherReads<-mean(dat_temp1[,fatherInd]) |
|
| 404 |
+ } |
|
| 405 |
+ else{
|
|
| 406 |
+ fatherReads<-0 |
|
| 407 |
+ } |
|
| 408 |
+ } else{
|
|
| 409 |
+ fatherReads<- "-" |
|
| 410 |
+ } |
|
| 411 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 412 |
+ if(length(motherInd)>1){
|
|
| 413 |
+ for(j in motherInd){
|
|
| 414 |
+ mothercount<-c(mothercount,mean(dat_temp1[,j])) |
|
| 415 |
+ } |
|
| 416 |
+ motherReads<-paste(mothercount,collapse=":") |
|
| 417 |
+ } |
|
| 418 |
+ else if(length(motherInd)==1){
|
|
| 419 |
+ motherReads<-mean(dat_temp1[,motherInd]) |
|
| 420 |
+ } |
|
| 421 |
+ else{
|
|
| 422 |
+ motherReads<-0 |
|
| 423 |
+ } |
|
| 424 |
+ } else{
|
|
| 425 |
+ motherReads <- "-" |
|
| 426 |
+ } |
|
| 427 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 428 |
+ if(length(probandInd)>1){
|
|
| 429 |
+ for(j in probandInd){
|
|
| 430 |
+ probandcount<-c(probandcount,mean(dat_temp1[,j])) |
|
| 431 |
+ } |
|
| 432 |
+ probandReads<-paste(probandcount,collapse=":") |
|
| 433 |
+ } |
|
| 434 |
+ else if(length(probandInd)==1){
|
|
| 435 |
+ probandReads<-mean(dat_temp1[,probandInd]) |
|
| 436 |
+ } |
|
| 437 |
+ else{
|
|
| 438 |
+ probandReads<-0 |
|
| 439 |
+ } |
|
| 440 |
+ } else{
|
|
| 441 |
+ probandReads <- "-" |
|
| 442 |
+ } |
|
| 443 |
+ |
|
| 444 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 445 |
+ if(length(siblingInd)>1){
|
|
| 446 |
+ for(j in siblingInd){
|
|
| 447 |
+ siblingcount<-c(siblingcount,mean(dat_temp1[,j])) |
|
| 448 |
+ } |
|
| 449 |
+ siblingReads<-paste(siblingcount,collapse=":") |
|
| 450 |
+ } |
|
| 451 |
+ else if(length(siblingInd)==1){
|
|
| 452 |
+ siblingReads<-mean(dat_temp1[,siblingInd]) |
|
| 453 |
+ } |
|
| 454 |
+ else{
|
|
| 455 |
+ siblingReads <- 0 |
|
| 456 |
+ } |
|
| 457 |
+ } |
|
| 458 |
+ else{
|
|
| 459 |
+ siblingReads<- "-" |
|
| 460 |
+ }' |
|
| 461 |
+ #motherReads<-dat_temp1[,motherInd] |
|
| 462 |
+ #probandReads<-dat_temp1[,probandInd] |
|
| 463 |
+ #if(is.na(siblingInd[1])==TRUE){
|
|
| 464 |
+ #siblingReads<-"-" |
|
| 465 |
+ #} |
|
| 466 |
+ #else{
|
|
| 467 |
+ #siblingReads<-dat_temp1[,siblingInd] |
|
| 468 |
+ #} |
|
| 469 |
+ } |
|
| 470 |
+ else if (nrow(dat_temp)==1){
|
|
| 471 |
+ |
|
| 472 |
+ #print(dim(dat_temp1)) |
|
| 473 |
+ fathercount<-c();mothercount<-c();probandcount<-c();siblingcount<-c() |
|
| 474 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 475 |
+ if(length(fatherInd)>1){
|
|
| 476 |
+ for(j in fatherInd){
|
|
| 477 |
+ fathercount<-c(fathercount,mean(dat_temp[,j])) |
|
| 478 |
+ } |
|
| 479 |
+ fatherReads<-paste(fathercount,collapse=":") |
|
| 480 |
+ } |
|
| 481 |
+ else if(length(fatherInd)==1){
|
|
| 482 |
+ fatherReads<-mean(dat_temp[,fatherInd]) |
|
| 483 |
+ } |
|
| 484 |
+ else{
|
|
| 485 |
+ fatherReads<-0 |
|
| 486 |
+ } |
|
| 487 |
+ }else{
|
|
| 488 |
+ fatherReads<-"-" |
|
| 489 |
+ } |
|
| 490 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 491 |
+ if(length(motherInd)>1){
|
|
| 492 |
+ for(j in motherInd){
|
|
| 493 |
+ mothercount<-c(mothercount,mean(dat_temp[,j])) |
|
| 494 |
+ } |
|
| 495 |
+ motherReads<-paste(mothercount,collapse=":") |
|
| 496 |
+ } |
|
| 497 |
+ else if(length(motherInd)==1){
|
|
| 498 |
+ motherReads<-mean(dat_temp[,motherInd]) |
|
| 499 |
+ } |
|
| 500 |
+ else{
|
|
| 501 |
+ motherReads<-0 |
|
| 502 |
+ } |
|
| 503 |
+ } else{
|
|
| 504 |
+ motherReads<-"-" |
|
| 505 |
+ } |
|
| 506 |
+ if(is.na(probandInd [1])==FALSE){
|
|
| 507 |
+ if(length(probandInd)>1){
|
|
| 508 |
+ for(j in probandInd){
|
|
| 509 |
+ probandcount<-c(probandcount,mean(dat_temp[,j])) |
|
| 510 |
+ } |
|
| 511 |
+ probandReads<-paste(probandcount,collapse=":") |
|
| 512 |
+ } |
|
| 513 |
+ else if(length(probandInd)==1){
|
|
| 514 |
+ probandReads<-mean(dat_temp[,probandInd]) |
|
| 515 |
+ } |
|
| 516 |
+ else{
|
|
| 517 |
+ probandReads<-0 |
|
| 518 |
+ } |
|
| 519 |
+ } else{
|
|
| 520 |
+ probandReads<-"-" |
|
| 521 |
+ } |
|
| 522 |
+ |
|
| 523 |
+ |
|
| 524 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 525 |
+ if(length(siblingInd)>1){
|
|
| 526 |
+ for(j in siblingInd){
|
|
| 527 |
+ siblingcount<-c(siblingcount,mean(dat_temp[,j])) |
|
| 528 |
+ } |
|
| 529 |
+ siblingReads<-paste(siblingcount,collapse=":") |
|
| 530 |
+ } |
|
| 531 |
+ else if(length(siblingInd)==1){
|
|
| 532 |
+ siblingReads<-mean(dat_temp[,siblingInd]) |
|
| 533 |
+ } |
|
| 534 |
+ else{
|
|
| 535 |
+ siblingReads<-0 |
|
| 536 |
+ } |
|
| 537 |
+ } |
|
| 538 |
+ else{
|
|
| 539 |
+ siblingReads<-"-" |
|
| 540 |
+ }' |
|
| 541 |
+ } |
|
| 542 |
+ else{
|
|
| 543 |
+ fatherReads<-"-" |
|
| 544 |
+ motherReads<-"-" |
|
| 545 |
+ probandReads<-"-" |
|
| 546 |
+ 'if(is.na(siblingInd[1])==TRUE){
|
|
| 547 |
+ siblingReads<-"-" |
|
| 548 |
+ } |
|
| 549 |
+ else{
|
|
| 550 |
+ siblingReads<-"-" |
|
| 551 |
+ }' |
|
| 552 |
+ } |
|
| 553 |
+ #gene<-overlap_ensemblgenes$SYMBOL |
|
| 554 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 555 |
+ gene<- as.character(gnsOverlapID) |
|
| 556 |
+ #ProbandGenes<-c() |
|
| 557 |
+ ProbandGenes<-paste(gene,"(",probandReads,")",sep="")
|
|
| 558 |
+ #ProbandGenes<-c(ProbandGenes,pasgene) |
|
| 559 |
+ ProbandTPM<-as.character(ProbandGenes) |
|
| 560 |
+ }else{
|
|
| 561 |
+ ProbandTPM <- "-" |
|
| 562 |
+ } |
|
| 563 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 564 |
+ FatherGenes<-paste(gene,"(",fatherReads,")",sep="")
|
|
| 565 |
+ FatherTPM<-as.character(FatherGenes) |
|
| 566 |
+ } |
|
| 567 |
+ else{
|
|
| 568 |
+ FatherTPM <- "-" |
|
| 569 |
+ } |
|
| 570 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 571 |
+ MotherGenes<-paste(gene,"(",motherReads,")",sep="")
|
|
| 572 |
+ #MotherGenes<-c(MotherGenes,pasgene) |
|
| 573 |
+ #} |
|
| 574 |
+ MotherTPM<-as.character(MotherGenes) |
|
| 575 |
+ } |
|
| 576 |
+ else{
|
|
| 577 |
+ MotherTPM <- "-" |
|
| 578 |
+ } |
|
| 579 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 580 |
+ siblingGenes<-paste(gene,"(",siblingReads,")",sep="")
|
|
| 581 |
+ #siblingGenes<-c(siblingGenes,pasgene) |
|
| 582 |
+ #} |
|
| 583 |
+ SiblingTPM<-as.character(siblingGenes) |
|
| 584 |
+ } |
|
| 585 |
+ else{
|
|
| 586 |
+ SiblingTPM<-"-" |
|
| 587 |
+ }' |
|
| 588 |
+ datGeneInfo<-data.frame(SVID=SVID,Probandexpression=ProbandTPM, |
|
| 589 |
+ Fatherexpression=FatherTPM,Motherexpression=MotherTPM) |
|
| 590 |
+ |
|
| 591 |
+ } |
|
| 592 |
+ else{
|
|
| 593 |
+ datGeneInfo<-data.frame(SVID=SVID,Probandexpression="-", |
|
| 594 |
+ Fatherexpression="-",Motherexpression="-") |
|
| 595 |
+ |
|
| 596 |
+ |
|
| 597 |
+ |
|
| 598 |
+ } |
|
| 599 |
+ |
|
| 600 |
+ #print(warnings()) |
|
| 601 |
+ |
|
| 602 |
+ return(datGeneInfo) |
|
| 603 |
+} |
|
| 604 |
+ |
|
| 605 |
+ |
|
| 606 |
+#' Extract Read counts for genes that are near SVs. |
|
| 607 |
+#' |
|
| 608 |
+#' @param gnsNonOverlap character. genes that are upstream |
|
| 609 |
+#' and/or downstream of SV. |
|
| 610 |
+#' @param SVID character. ID of the SVs. |
|
| 611 |
+#' @param RNASeqData character. Expression of the genes. |
|
| 612 |
+#' @param pattern_Proband character. Pattern to identify the proband reads. |
|
| 613 |
+#' @param pattern_Father character. Pattern to identify the father reads. |
|
| 614 |
+#' @param pattern_Mother character. Pattern to identify the mother reads. |
|
| 615 |
+#' @return Text or Dataframe containing TPM read counts of genes in the family. |
|
| 616 |
+#' @examples |
|
| 617 |
+#' RNASeqDir = system.file("extdata", package="nanotatoR")
|
|
| 618 |
+#' returnMethod="dataFrame" |
|
| 619 |
+#' datRNASeq <- RNAseqcombine(RNASeqDir = RNASeqDir, |
|
| 620 |
+#' returnMethod = returnMethod) |
|
| 621 |
+#' gnsNonOverlap <- c("DDX11L1", "MIR1302-2HG", "OR4G4P")
|
|
| 622 |
+#' SVID = 397 |
|
| 623 |
+#' datgnnonovrlap <- nonOverlapRNAseq(gnsNonOverlap = gnsNonOverlap, |
|
| 624 |
+#' SVID = SVID, RNASeqData = datRNASeq, |
|
| 625 |
+#' pattern_Proband = "*_P_*") |
|
| 626 |
+#' @importFrom stats na.omit |
|
| 627 |
+#' @export |
|
| 628 |
+nonOverlapRNAseq<-function(gnsNonOverlap,SVID,RNASeqData, |
|
| 629 |
+ pattern_Proband=NA,pattern_Mother=NA, |
|
| 630 |
+ pattern_Father=NA){
|
|
| 631 |
+ ##Biomart annotation |
|
| 632 |
+ ###Checking if the input is empty; else if not empty add |
|
| 633 |
+ ###expression values for each genes |
|
| 634 |
+ datGeneInfo<-data.frame() |
|
| 635 |
+ SVID=SVID |
|
| 636 |
+ ###Extracting the index for the the parents |
|
| 637 |
+ if(is.na(unique(as.character(pattern_Father))) == FALSE){
|
|
| 638 |
+ fatherInd<-grep(unique(as.character(pattern_Father)),names(RNASeqData)) |
|
| 639 |
+ } else{fatherInd<- NA}
|
|
| 640 |
+ if(is.na(unique(as.character(pattern_Mother))) == FALSE){
|
|
| 641 |
+ motherInd<-grep(unique(as.character(pattern_Mother)),names(RNASeqData)) |
|
| 642 |
+ } else{motherInd<- NA}
|
|
| 643 |
+ if(is.na(unique(as.character(pattern_Proband))) == FALSE){
|
|
| 644 |
+ probandInd<-grep(unique(as.character(pattern_Proband)),names(RNASeqData)) |
|
| 645 |
+ } else{probandInd<- NA}
|
|
| 646 |
+ |
|
| 647 |
+ ##Checking for sibling |
|
| 648 |
+ 'if(is.na(pattern_Sibling)==FALSE){
|
|
| 649 |
+ siblingInd<-grep(pattern_Sibling,names(RNASeqData)) |
|
| 650 |
+ } |
|
| 651 |
+ else{
|
|
| 652 |
+ siblingInd<-NA |
|
| 653 |
+ }' |
|
| 654 |
+ |
|
| 655 |
+ gene<-c() |
|
| 656 |
+ gnsname<-as.character(RNASeqData$GeneName) |
|
| 657 |
+ pasgnsname<-pasgnovlap<-paste("^",as.character(gnsname),"$",sep="")
|
|
| 658 |
+ 'nonoverlap_ensemblgenes = select(EnsDb.Hsapiens.v79, gnsNonOverlap, |
|
| 659 |
+ c("GENEID","GENENAME"), "SYMBOL")
|
|
| 660 |
+ gnsnonOverlapID<-as.character(nonoverlap_ensemblgenes$GENEID)' |
|
| 661 |
+ ###Extracting Reads |
|
| 662 |
+ ### |
|
| 663 |
+ ###Genes Names Extraction |
|
| 664 |
+ gnsnonOverlapID<- as.character(gnsNonOverlap) |
|
| 665 |
+ if(length(gnsnonOverlapID)>1){
|
|
| 666 |
+ #datGeneInfoTemp<-data.frame() |
|
| 667 |
+ fatherReads<-c() |
|
| 668 |
+ motherReads<-c() |
|
| 669 |
+ probandReads<-c() |
|
| 670 |
+ #siblingReads<-c() |
|
| 671 |
+ |
|
| 672 |
+ for (ki in 1:length(gnsnonOverlapID)){
|
|
| 673 |
+ pasgnnonovlap<-paste("^",as.character(gnsnonOverlapID[ki]),"$",sep="")
|
|
| 674 |
+ gg<-grep(pasgnnonovlap,pasgnsname,fixed=TRUE) |
|
| 675 |
+ dat_temp<-RNASeqData[gg,] |
|
| 676 |
+ if(nrow(dat_temp)>1){
|
|
| 677 |
+ dat_temp_1<-apply(dat_temp[,3:ncol(dat_temp)],2,mean) |
|
| 678 |
+ #dat_temp_1<-data.frame(dat_temp_1) |
|
| 679 |
+ dat_temp1<-dat_temp[1,] |
|
| 680 |
+ dat_temp1[,3:ncol(dat_temp)]<-dat_temp_1 |
|
| 681 |
+ #print(dim(dat_temp1)) |
|
| 682 |
+ fathercount<-c();mothercount<-c();probandcount<-c();siblingcount<-c() |
|
| 683 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 684 |
+ if(length(fatherInd)>1){
|
|
| 685 |
+ for(j in fatherInd){
|
|
| 686 |
+ fathercount<-c(fathercount,dat_temp1[,j]) |
|
| 687 |
+ } |
|
| 688 |
+ fatherReads<-c(fatherReads,paste(fathercount,collapse=":")) |
|
| 689 |
+ } |
|
| 690 |
+ else if(length(fatherInd)==1){
|
|
| 691 |
+ fatherReads<-c(fatherReads,dat_temp1[,fatherInd]) |
|
| 692 |
+ } |
|
| 693 |
+ else{
|
|
| 694 |
+ fatherReads<-c(fatherReads,0) |
|
| 695 |
+ } |
|
| 696 |
+ } |
|
| 697 |
+ else{
|
|
| 698 |
+ fatherReads<-c(fatherReads,"-") |
|
| 699 |
+ } |
|
| 700 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 701 |
+ if(length(motherInd)>1){
|
|
| 702 |
+ for(j in motherInd){
|
|
| 703 |
+ mothercount<-c(mothercount,dat_temp1[,j]) |
|
| 704 |
+ } |
|
| 705 |
+ motherReads<-c(motherReads,paste(mothercount,collapse=":")) |
|
| 706 |
+ } |
|
| 707 |
+ else if(length(motherInd)==1){
|
|
| 708 |
+ motherReads<-c(motherReads,dat_temp1[,motherInd]) |
|
| 709 |
+ } |
|
| 710 |
+ else{
|
|
| 711 |
+ motherReads<-c(motherReads,0) |
|
| 712 |
+ } |
|
| 713 |
+ } else{
|
|
| 714 |
+ motherReads<-c(motherReads, "-") |
|
| 715 |
+ } |
|
| 716 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 717 |
+ if(length(probandInd)>1){
|
|
| 718 |
+ for(j in probandInd){
|
|
| 719 |
+ probandcount<-c(probandcount,dat_temp1[,j]) |
|
| 720 |
+ } |
|
| 721 |
+ probandReads<-c(probandReads,paste(probandcount,collapse=":")) |
|
| 722 |
+ } |
|
| 723 |
+ else if(length(probandInd)==1){
|
|
| 724 |
+ probandReads<-c(probandReads,dat_temp1[,probandInd]) |
|
| 725 |
+ } |
|
| 726 |
+ else{
|
|
| 727 |
+ probandReads<-c(probandReads,0) |
|
| 728 |
+ } |
|
| 729 |
+ } else{
|
|
| 730 |
+ probandReads<-c(probandReads, "-") |
|
| 731 |
+ } |
|
| 732 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 733 |
+ if(length(siblingInd)>1){
|
|
| 734 |
+ for(j in siblingInd){
|
|
| 735 |
+ siblingcount<-c(siblingcount,dat_temp1[,j]) |
|
| 736 |
+ } |
|
| 737 |
+ siblingReads<-c(siblingReads,paste(siblingcount,collapse=":")) |
|
| 738 |
+ } |
|
| 739 |
+ else if(length(siblingInd)==1){
|
|
| 740 |
+ siblingReads<-c(siblingReads,dat_temp1[,siblingInd]) |
|
| 741 |
+ } |
|
| 742 |
+ else{
|
|
| 743 |
+ siblingReads<-c(siblingReads,0) |
|
| 744 |
+ } |
|
| 745 |
+ } |
|
| 746 |
+ else{
|
|
| 747 |
+ siblingReads<-c(siblingReads,"-") |
|
| 748 |
+ }' |
|
| 749 |
+ } |
|
| 750 |
+ else if (nrow(dat_temp)==1) {
|
|
| 751 |
+ #print(dim(dat_temp1)) |
|
| 752 |
+ fathercount<-c();mothercount<-c();probandcount<-c(); |
|
| 753 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 754 |
+ if(length(fatherInd)>1){
|
|
| 755 |
+ for(j in fatherInd){
|
|
| 756 |
+ fathercount<-c(fathercount,dat_temp[,j]) |
|
| 757 |
+ } |
|
| 758 |
+ fatherReads<-c(fatherReads,paste(fathercount,collapse=":")) |
|
| 759 |
+ } |
|
| 760 |
+ else if(length(fatherInd)==1){
|
|
| 761 |
+ fatherReads<-c(fatherReads,dat_temp[,fatherInd]) |
|
| 762 |
+ } |
|
| 763 |
+ else{
|
|
| 764 |
+ fatherReads<-c(fatherReads,0) |
|
| 765 |
+ } |
|
| 766 |
+ } |
|
| 767 |
+ else{
|
|
| 768 |
+ fatherReads<-c(fatherReads,"-") |
|
| 769 |
+ } |
|
| 770 |
+ |
|
| 771 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 772 |
+ if(length(motherInd)>1){
|
|
| 773 |
+ for(j in motherInd){
|
|
| 774 |
+ mothercount<-c(mothercount,dat_temp[,j]) |
|
| 775 |
+ } |
|
| 776 |
+ motherReads<-c(motherReads,paste(mothercount,collapse=":")) |
|
| 777 |
+ } |
|
| 778 |
+ else if(length(motherInd)==1){
|
|
| 779 |
+ motherReads<-c(motherReads,dat_temp[,motherInd]) |
|
| 780 |
+ } |
|
| 781 |
+ else{
|
|
| 782 |
+ motherReads<-c(motherReads,0) |
|
| 783 |
+ } |
|
| 784 |
+ } |
|
| 785 |
+ else{
|
|
| 786 |
+ motherReads<-c(motherReads,"-") |
|
| 787 |
+ } |
|
| 788 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 789 |
+ if(length(probandInd)>1){
|
|
| 790 |
+ for(j in probandInd){
|
|
| 791 |
+ probandcount<-c(probandcount,dat_temp[,j]) |
|
| 792 |
+ } |
|
| 793 |
+ probandReads<-c(probandReads,paste(probandcount,collapse=":")) |
|
| 794 |
+ } |
|
| 795 |
+ else if(length(probandInd)==1){
|
|
| 796 |
+ probandReads<-c(probandReads,dat_temp[,probandInd]) |
|
| 797 |
+ } |
|
| 798 |
+ else{
|
|
| 799 |
+ probandReads<-c(probandReads,0) |
|
| 800 |
+ } |
|
| 801 |
+ } |
|
| 802 |
+ else{
|
|
| 803 |
+ probandReads <- c(probandReads,"-") |
|
| 804 |
+ } |
|
| 805 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 806 |
+ if(length(siblingInd)>1){
|
|
| 807 |
+ for(j in siblingInd){
|
|
| 808 |
+ siblingcount<-c(siblingcount,dat_temp[,j]) |
|
| 809 |
+ } |
|
| 810 |
+ siblingReads<-c(siblingReads,paste(siblingcount,collapse=":")) |
|
| 811 |
+ } |
|
| 812 |
+ else if(length(siblingInd)==1){
|
|
| 813 |
+ siblingReads<-c(siblingReads,dat_temp[,siblingInd]) |
|
| 814 |
+ } |
|
| 815 |
+ else{
|
|
| 816 |
+ siblingReads<-c(siblingReads,0) |
|
| 817 |
+ } |
|
| 818 |
+ } |
|
| 819 |
+ else{
|
|
| 820 |
+ siblingReads<-c(siblingReads,"-") |
|
| 821 |
+ }' |
|
| 822 |
+ } |
|
| 823 |
+ else{
|
|
| 824 |
+ fatherReads<-c(fatherReads,"-") |
|
| 825 |
+ motherReads<-c(motherReads,"-") |
|
| 826 |
+ probandReads<-c(probandReads,"-") |
|
| 827 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 828 |
+ siblingReads<-c(siblingReads,"-") |
|
| 829 |
+ } |
|
| 830 |
+ else{
|
|
| 831 |
+ siblingReads<-c(siblingReads,"-") |
|
| 832 |
+ }' |
|
| 833 |
+ } |
|
| 834 |
+ gene<-c(gene,as.character(gnsnonOverlapID[ki])) |
|
| 835 |
+ } |
|
| 836 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 837 |
+ ProbandGenes<-c() |
|
| 838 |
+ |
|
| 839 |
+ for(ii in 1:length(gene)){
|
|
| 840 |
+ pasgene<-paste(gene[ii],"(",probandReads[ii],")",sep="")
|
|
| 841 |
+ ProbandGenes<-c(ProbandGenes,pasgene) |
|
| 842 |
+ } |
|
| 843 |
+ ProbandTPM<-paste(ProbandGenes,collapse=";") |
|
| 844 |
+ } else{
|
|
| 845 |
+ ProbandTPM<-"-" |
|
| 846 |
+ } |
|
| 847 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 848 |
+ FatherGenes<-c() |
|
| 849 |
+ for(ii in 1:length(gene)){
|
|
| 850 |
+ pasgene<-paste(gene[ii],"(",fatherReads[ii],")",sep="")
|
|
| 851 |
+ FatherGenes<-c(FatherGenes,pasgene) |
|
| 852 |
+ } |
|
| 853 |
+ FatherTPM<-paste(FatherGenes,collapse=";") |
|
| 854 |
+ }else{
|
|
| 855 |
+ FatherTPM<-"-" |
|
| 856 |
+ } |
|
| 857 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 858 |
+ MotherGenes<-c() |
|
| 859 |
+ for(ii in 1:length(gene)){
|
|
| 860 |
+ pasgene<-paste(gene[ii],"(",motherReads[ii],")",sep="")
|
|
| 861 |
+ MotherGenes<-c(MotherGenes,pasgene) |
|
| 862 |
+ } |
|
| 863 |
+ MotherTPM<-paste(MotherGenes,collapse=";") |
|
| 864 |
+ }else{
|
|
| 865 |
+ MotherTPM<-"-" |
|
| 866 |
+ } |
|
| 867 |
+ |
|
| 868 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 869 |
+ siblingGenes<-c() |
|
| 870 |
+ for(ii in 1:length(gene)){
|
|
| 871 |
+ pasgene<-paste(gene[ii],"(",siblingReads[ii],")",sep="")
|
|
| 872 |
+ siblingGenes<-c(siblingGenes,pasgene) |
|
| 873 |
+ } |
|
| 874 |
+ SiblingTPM<-paste(siblingGenes,collapse=";") |
|
| 875 |
+ } |
|
| 876 |
+ else{
|
|
| 877 |
+ SiblingTPM<-"-" |
|
| 878 |
+ }' |
|
| 879 |
+ datGeneInfo<-data.frame(SVID=SVID,ProbandTPM=ProbandTPM, |
|
| 880 |
+ FatherTPM=FatherTPM,MotherTPM=MotherTPM) |
|
| 881 |
+ } |
|
| 882 |
+ else if(length(gnsnonOverlapID)==1){
|
|
| 883 |
+ pasgnnonovlap<-paste("^",as.character(gnsnonOverlapID),"$",sep="")
|
|
| 884 |
+ gg<-grep(pasgnnonovlap,pasgnsname,fixed=TRUE) |
|
| 885 |
+ #gg<-grep(pasgnnonovlap,gnsname,fixed=TRUE) |
|
| 886 |
+ dat_temp<-RNASeqData[gg,] |
|
| 887 |
+ |
|
| 888 |
+ if(nrow(dat_temp)>1){
|
|
| 889 |
+ dat_temp_1<-apply(dat_temp[,3:ncol(dat_temp)],2,mean) |
|
| 890 |
+ #dat_temp_1<-data.frame(dat_temp_1) |
|
| 891 |
+ dat_temp1<-dat_temp[1,] |
|
| 892 |
+ dat_temp1[,3:ncol(dat_temp)]<-dat_temp_1 |
|
| 893 |
+ #print(dim(dat_temp1)) |
|
| 894 |
+ fathercount<-c();mothercount<-c();probandcount<-c(); |
|
| 895 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 896 |
+ if(length(fatherInd)>1){
|
|
| 897 |
+ for(j in fatherInd){
|
|
| 898 |
+ fathercount<-c(fathercount,dat_temp1[,j]) |
|
| 899 |
+ } |
|
| 900 |
+ fatherReads<-paste(fathercount,collapse=":") |
|
| 901 |
+ } |
|
| 902 |
+ else if(length(fatherInd)==1){
|
|
| 903 |
+ fatherReads<-dat_temp1[,fatherInd] |
|
| 904 |
+ } |
|
| 905 |
+ else{
|
|
| 906 |
+ fatherReads<-0 |
|
| 907 |
+ } |
|
| 908 |
+ } |
|
| 909 |
+ else{
|
|
| 910 |
+ fatherReads<- "-" |
|
| 911 |
+ } |
|
| 912 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 913 |
+ if(length(motherInd)>1){
|
|
| 914 |
+ for(j in motherInd){
|
|
| 915 |
+ mothercount<-c(mothercount,dat_temp1[,j]) |
|
| 916 |
+ } |
|
| 917 |
+ motherReads<-paste(mothercount,collapse=":") |
|
| 918 |
+ } |
|
| 919 |
+ else if(length(motherInd)==1){
|
|
| 920 |
+ motherReads<-dat_temp1[,motherInd] |
|
| 921 |
+ } |
|
| 922 |
+ else{
|
|
| 923 |
+ motherReads<-0 |
|
| 924 |
+ } |
|
| 925 |
+ } |
|
| 926 |
+ else{
|
|
| 927 |
+ motherReads<- "-" |
|
| 928 |
+ } |
|
| 929 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 930 |
+ if(length(probandInd)>1){
|
|
| 931 |
+ for(j in probandInd){
|
|
| 932 |
+ probandcount<-c(probandcount,dat_temp1[,j]) |
|
| 933 |
+ } |
|
| 934 |
+ probandReads<-paste(probandcount,collapse=":") |
|
| 935 |
+ } |
|
| 936 |
+ else if(length(probandInd)==1){
|
|
| 937 |
+ probandReads<-dat_temp1[,probandInd] |
|
| 938 |
+ } |
|
| 939 |
+ else{
|
|
| 940 |
+ probandReads<-0 |
|
| 941 |
+ } |
|
| 942 |
+ } |
|
| 943 |
+ else{
|
|
| 944 |
+ probandReads<- "-" |
|
| 945 |
+ } |
|
| 946 |
+ |
|
| 947 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 948 |
+ if(length(siblingInd)>1){
|
|
| 949 |
+ for(j in siblingInd){
|
|
| 950 |
+ siblingcount<-c(siblingcount,dat_temp1[,j]) |
|
| 951 |
+ } |
|
| 952 |
+ siblingReads<-paste(siblingcount,collapse=":") |
|
| 953 |
+ } |
|
| 954 |
+ else if(length(siblingInd)==1){
|
|
| 955 |
+ siblingReads<-dat_temp1[,siblingInd] |
|
| 956 |
+ } |
|
| 957 |
+ else{
|
|
| 958 |
+ siblingReads <- 0 |
|
| 959 |
+ } |
|
| 960 |
+ } |
|
| 961 |
+ else{
|
|
| 962 |
+ siblingReads<- "-" |
|
| 963 |
+ }' |
|
| 964 |
+ #motherReads<-dat_temp1[,motherInd] |
|
| 965 |
+ #probandReads<-dat_temp1[,probandInd] |
|
| 966 |
+ #if(is.na(siblingInd[1])==TRUE){
|
|
| 967 |
+ #siblingReads<-"-" |
|
| 968 |
+ #} |
|
| 969 |
+ #else{
|
|
| 970 |
+ #siblingReads<-dat_temp1[,siblingInd] |
|
| 971 |
+ #} |
|
| 972 |
+ } |
|
| 973 |
+ else if (nrow(dat_temp)==1){
|
|
| 974 |
+ |
|
| 975 |
+ #print(dim(dat_temp1)) |
|
| 976 |
+ fathercount<-c();mothercount<-c();probandcount<-c(); |
|
| 977 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 978 |
+ if(length(fatherInd)>1){
|
|
| 979 |
+ for(j in fatherInd){
|
|
| 980 |
+ fathercount<-c(fathercount,dat_temp[,j]) |
|
| 981 |
+ } |
|
| 982 |
+ fatherReads<-paste(fathercount,collapse=":") |
|
| 983 |
+ } |
|
| 984 |
+ else if(length(fatherInd)==1){
|
|
| 985 |
+ fatherReads<-dat_temp[,fatherInd] |
|
| 986 |
+ } |
|
| 987 |
+ else{
|
|
| 988 |
+ fatherReads<-0 |
|
| 989 |
+ } |
|
| 990 |
+ } |
|
| 991 |
+ else{
|
|
| 992 |
+ fatherReads<- "-" |
|
| 993 |
+ } |
|
| 994 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 995 |
+ if(length(motherInd)>1){
|
|
| 996 |
+ for(j in motherInd){
|
|
| 997 |
+ mothercount<-c(mothercount,dat_temp[,j]) |
|
| 998 |
+ } |
|
| 999 |
+ motherReads<-paste(mothercount,collapse=":") |
|
| 1000 |
+ } |
|
| 1001 |
+ else if(length(motherInd)==1){
|
|
| 1002 |
+ motherReads<-dat_temp[,motherInd] |
|
| 1003 |
+ } |
|
| 1004 |
+ else{
|
|
| 1005 |
+ motherReads<-0 |
|
| 1006 |
+ } |
|
| 1007 |
+ } |
|
| 1008 |
+ else{
|
|
| 1009 |
+ motherReads<- "-" |
|
| 1010 |
+ } |
|
| 1011 |
+ if(is.na(probandInd)==FALSE){
|
|
| 1012 |
+ if(length(probandInd)>1){
|
|
| 1013 |
+ for(j in probandInd){
|
|
| 1014 |
+ probandcount<-c(probandcount,dat_temp[,j]) |
|
| 1015 |
+ } |
|
| 1016 |
+ probandReads<-paste(probandcount,collapse=":") |
|
| 1017 |
+ } |
|
| 1018 |
+ else if(length(probandInd)==1){
|
|
| 1019 |
+ probandReads<-dat_temp[,probandInd] |
|
| 1020 |
+ } |
|
| 1021 |
+ else{
|
|
| 1022 |
+ probandReads<-0 |
|
| 1023 |
+ } |
|
| 1024 |
+ } |
|
| 1025 |
+ else{
|
|
| 1026 |
+ probandReads<- "-" |
|
| 1027 |
+ } |
|
| 1028 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 1029 |
+ if(length(siblingInd)>1){
|
|
| 1030 |
+ for(j in siblingInd){
|
|
| 1031 |
+ siblingcount<-c(siblingcount,dat_temp[,j]) |
|
| 1032 |
+ } |
|
| 1033 |
+ siblingReads<-paste(siblingcount,collapse=":") |
|
| 1034 |
+ } |
|
| 1035 |
+ else if(length(siblingInd)==1){
|
|
| 1036 |
+ siblingReads<-dat_temp[,siblingInd] |
|
| 1037 |
+ } |
|
| 1038 |
+ else{
|
|
| 1039 |
+ siblingReads<-0 |
|
| 1040 |
+ } |
|
| 1041 |
+ } |
|
| 1042 |
+ else{
|
|
| 1043 |
+ siblingReads<-"-" |
|
| 1044 |
+ }' |
|
| 1045 |
+ } |
|
| 1046 |
+ else{
|
|
| 1047 |
+ fatherReads<-"-" |
|
| 1048 |
+ motherReads<-"-" |
|
| 1049 |
+ probandReads<-"-" |
|
| 1050 |
+ 'if(is.na(siblingInd[1])==TRUE){
|
|
| 1051 |
+ siblingReads<-"-" |
|
| 1052 |
+ } |
|
| 1053 |
+ else{
|
|
| 1054 |
+ siblingReads<-"-" |
|
| 1055 |
+ }' |
|
| 1056 |
+ } |
|
| 1057 |
+ gene<-c(gene,as.character(gnsnonOverlapID)) |
|
| 1058 |
+ if(is.na(probandInd[1])==FALSE){
|
|
| 1059 |
+ ProbandGenes<-c() |
|
| 1060 |
+ ProbandGenes<-paste(gene,"(",probandReads,")",sep="")
|
|
| 1061 |
+ #ProbandGenes<-c(ProbandGenes,pasgene) |
|
| 1062 |
+ #} |
|
| 1063 |
+ ProbandTPM<-as.character(ProbandGenes) |
|
| 1064 |
+ }else{
|
|
| 1065 |
+ ProbandTPM<-"-" |
|
| 1066 |
+ } |
|
| 1067 |
+ if(is.na(fatherInd[1])==FALSE){
|
|
| 1068 |
+ FatherGenes<-c() |
|
| 1069 |
+ FatherGenes<-paste(gene,"(",fatherReads,")",sep="")
|
|
| 1070 |
+ FatherTPM<-as.character(FatherGenes) |
|
| 1071 |
+ }else{
|
|
| 1072 |
+ FatherTPM <- "-" |
|
| 1073 |
+ } |
|
| 1074 |
+ if(is.na(motherInd[1])==FALSE){
|
|
| 1075 |
+ MotherGenes<-c() |
|
| 1076 |
+ MotherGenes<-paste(gene,"(",motherReads,")",sep="")
|
|
| 1077 |
+ MotherTPM<-as.character(MotherGenes) |
|
| 1078 |
+ }else{
|
|
| 1079 |
+ MotherTPM <- "-" |
|
| 1080 |
+ } |
|
| 1081 |
+ |
|
| 1082 |
+ 'if(is.na(siblingInd[1])==FALSE){
|
|
| 1083 |
+ siblingGenes<-paste(gene,"(",siblingReads,")",sep="")
|
|
| 1084 |
+ #siblingGenes<-c(siblingGenes,pasgene) |
|
| 1085 |
+ #} |
|
| 1086 |
+ SiblingTPM<-as.character(siblingGenes) |
|
| 1087 |
+ } |
|
| 1088 |
+ else{
|
|
| 1089 |
+ SiblingTPM<-"-" |
|
| 1090 |
+ }' |
|
| 1091 |
+ |
|
| 1092 |
+ datGeneInfo<-data.frame(SVID=SVID,ProbandTPM=ProbandTPM, |
|
| 1093 |
+ FatherTPM=FatherTPM,MotherTPM=MotherTPM) |
|
| 1094 |
+ } |
|
| 1095 |
+ else{
|
|
| 1096 |
+ datGeneInfo<-data.frame(SVID=SVID,ProbandTPM="-", |
|
| 1097 |
+ FatherTPM="-",MotherTPM="-") |
|
| 1098 |
+ |
|
| 1099 |
+ } |
|
| 1100 |
+ |
|
| 1101 |
+ return(datGeneInfo) |
|
| 1102 |
+} |
|
| 1103 |
+ |
|
| 1104 |
+#' Extract Read counts for genes that are near |
|
| 1105 |
+#' or overalapping SVs. |
|
| 1106 |
+#' |
|
| 1107 |
+#' @param input_fmt_SV character. genes that are upstream |
|
| 1108 |
+#' and/or downstream of SV. input_fmt_RNASeq |
|
| 1109 |
+#' @param input_fmt_RNASeq character. input format of RNASEQ data. |
|
| 1110 |
+#' Text or dataframe. |
|
| 1111 |
+#' @param smapdata dataframe. smap data in dataframe format. |
|
| 1112 |
+#' @param smappath character. smap path. |
|
| 1113 |
+#' @param RNASeqPATH character. RNASEQ path. |
|
| 1114 |
+#' @param outputfmt character. Output format choice dataframe or text. |
|
| 1115 |
+#' @param RNASeqData character. Expression of the genes. |
|
| 1116 |
+#' @param pattern_Proband character. Pattern to identify the proband reads. |
|
| 1117 |
+#' @param pattern_Father character. Pattern to identify the father reads. |
|
| 1118 |
+#' @param pattern_Mother character. Pattern to identify the mother reads. |
|
| 1119 |
+#' @param EnzymeType character. Enzyme used. option "Dual" or "DLE". |
|
| 1120 |
+#' @return Text or Dataframe containing TPM read counts of genes in the family. |
|
| 1121 |
+#' @examples |
|
| 1122 |
+#' \dontrun{
|
|
| 1123 |
+#' RNASeqDir = system.file("extdata", package="nanotatoR")
|
|
| 1124 |
+#' returnMethod="dataFrame" |
|
| 1125 |
+#' datRNASeq <- RNAseqcombine(RNASeqDir = RNASeqDir, |
|
| 1126 |
+#' returnMethod = returnMethod) |
|
| 1127 |
+#' smapName="NA12878_DLE1_VAP_solo5.smap" |
|
| 1128 |
+#' smap = system.file("extdata", smapName, package="nanotatoR")
|
|
| 1129 |
+#' datcomp<-overlapnearestgeneSearch(smap = smap, |
|
| 1130 |
+#' bed=bedFile, inputfmtBed = "BED", outpath, |
|
| 1131 |
+#' n = 3, returnMethod_bedcomp = c("dataFrame"),
|
|
| 1132 |
+#' input_fmt_SV = "Text", |
|
| 1133 |
+#' EnzymeType = "SVMerge", |
|
| 1134 |
+#' bperrorindel = 3000, |
|
| 1135 |
+#' bperrorinvtrans = 50000) |
|
| 1136 |
+#' datRNASeq1 <- SVexpression( |
|
| 1137 |
+#' input_fmt_SV=c("dataFrame"),
|
|
| 1138 |
+# smapdata = datcomp, |
|
| 1139 |
+#' input_fmt_RNASeq=c("dataFrame"),
|
|
| 1140 |
+#' RNASeqData = datRNASeq, |
|
| 1141 |
+#' outputfmt=c("datFrame"),
|
|
| 1142 |
+#' pattern_Proband = "*_P_*", EnzymeType = c("SVMerge"))
|
|
| 1143 |
+ |
|
| 1144 |
+#'} |
|
| 1145 |
+#' @importFrom stats na.omit |
|
| 1146 |
+#' @export |
|
| 1147 |
+SVexpression_duo_trio <-function(input_fmt_SV=c("Text","dataFrame"),
|
|
| 1148 |
+ smapdata,smappath, |
|
| 1149 |
+ input_fmt_RNASeq=c("Text","dataFrame"),
|
|
| 1150 |
+ RNASeqData,RNASeqPATH,outputfmt=c("Text","datFrame"),
|
|
| 1151 |
+ pattern_Proband=NA,pattern_Mother=NA,pattern_Father=NA, |
|
| 1152 |
+ EnzymeType = c("SVMerge", "SE")){
|
|
| 1153 |
+ ###RNASEQ Analysis data |
|
| 1154 |
+ if(input_fmt_RNASeq=="dataFrame"){
|
|
| 1155 |
+ RNASeqData=RNASeqData |
|
| 1156 |
+ } |
|
| 1157 |
+ else if(input_fmt_RNASeq=="Text"){
|
|
| 1158 |
+ RNASeqData=read.csv(RNASeqPATH) |
|
| 1159 |
+ } |
|
| 1160 |
+ else{
|
|
| 1161 |
+ stop("Input format for RNASeq Data Incorrect")
|
|
| 1162 |
+ } |
|
| 1163 |
+ |
|
| 1164 |
+ if(input_fmt_SV=="dataFrame"){
|
|
| 1165 |
+ smapdata = smapdata |
|
| 1166 |
+ if(EnzymeType == "SVMerge"){
|
|
| 1167 |
+ #smapdata <- readSMap(smap, input_fmt_smap = "Text") |
|
| 1168 |
+ SVID<-smapdata$SVIndex |
|
| 1169 |
+ } |
|
| 1170 |
+ else{
|
|
| 1171 |
+ #smapdata <- readSMap_DLE(smap, input_fmt_smap) |
|
| 1172 |
+ SVID<-smapdata$SmapEntryID |
|
| 1173 |
+ } |
|
| 1174 |
+ } |
|
| 1175 |
+ else if(input_fmt_SV=="Text"){
|
|
| 1176 |
+ if(EnzymeType == "SVMerge"){
|
|
| 1177 |
+ smapdata <- readSMap(smappath, input_fmt_smap = "Text") |
|
| 1178 |
+ SVID<-smapdata$SVIndex |
|
| 1179 |
+ } |
|
| 1180 |
+ else{
|
|
| 1181 |
+ smapdata <- readSMap_DLE(smappath, input_fmt_smap = "Text") |
|
| 1182 |
+ SVID<-smapdata$SmapEntryID |
|
| 1183 |
+ } |
|
| 1184 |
+ } |
|
| 1185 |
+ else{
|
|
| 1186 |
+ stop("Input format for SMAP Incorrect")
|
|
| 1187 |
+ } |
|
| 1188 |
+ ##Extracting Data |
|
| 1189 |
+ overlapgenes<-str_trim(smapdata$OverlapGenes_strand_perc) |
|
| 1190 |
+ |
|
| 1191 |
+ SVID<-smapdata$SVIndex |
|
| 1192 |
+ dataOverLap<-data.frame(matrix(nrow=nrow(smapdata),ncol=4)) |
|
| 1193 |
+ ##Extracting Overlapped Genes |
|
| 1194 |
+ #dataOverLap<-data.frame(matrix(nrow=10,ncol=5)) |
|
| 1195 |
+ names(dataOverLap)<-c("SVID","OverlapProbandTPM",
|
|
| 1196 |
+ "OverlapFatherTPM","OverlapMotherTPM") |
|
| 1197 |
+ print("###OverlapGenes###")
|
|
| 1198 |
+ for(kk in 1:length(overlapgenes)){
|
|
| 1199 |
+ #print(kk) |
|
| 1200 |
+ #for(kk in 1:10){
|
|
| 1201 |
+ #print(kk) |
|
| 1202 |
+ datOverLap<-data.frame() |
|
| 1203 |
+ print(paste("kk:",kk,sep=""))
|
|
| 1204 |
+ svID<-as.character(SVID[kk]) |
|
| 1205 |
+ if(length(grep(";",overlapgenes[kk]))>=1){
|
|
| 1206 |
+ st1<-strsplit(as.character(overlapgenes[kk]), |
|
| 1207 |
+ split = ";") |
|
| 1208 |
+ sttemp<-as.character(st1[[1]]) |
|
| 1209 |
+ #print("1")
|
|
| 1210 |
+ gns_overlap<-c() |
|
| 1211 |
+ for (tt in 1:length(sttemp)){
|
|
| 1212 |
+ gn_temp<-strsplit(sttemp[tt],split="\\(")
|
|
| 1213 |
+ gns_overlap<-c(gns_overlap,as.character(gn_temp[[1]][1])) |
|
| 1214 |
+ } |
|
| 1215 |
+ |
|
| 1216 |
+ datOverLap<-OverlapRNAseq(gnsOverlap = as.character(gns_overlap), |
|
| 1217 |
+ SVID = svID,RNASeqData = RNASeqData, |
|
| 1218 |
+ pattern_Proband=pattern_Proband,pattern_Mother=pattern_Mother, |
|
| 1219 |
+ pattern_Father=pattern_Father) |
|
| 1220 |
+ |
|
| 1221 |
+ |
|
| 1222 |
+ }else if (length(grep("\\(",as.character(overlapgenes[kk])))>=1){
|
|
| 1223 |
+ #print("2")
|
|
| 1224 |
+ gnsOverlap<-strsplit(as.character(overlapgenes[kk]),split="\\(")[[1]][1]
|
|
| 1225 |
+ |
|
| 1226 |
+ datOverLap<-OverlapRNAseq(gnsOverlap = as.character(gnsOverlap), |
|
| 1227 |
+ SVID = svID,RNASeqData = RNASeqData, |
|
| 1228 |
+ pattern_Proband = pattern_Proband, |
|
| 1229 |
+ pattern_Mother = pattern_Mother, |
|
| 1230 |
+ pattern_Father = pattern_Father) |
|
| 1231 |
+ |
|
| 1232 |
+ }else{
|
|
| 1233 |
+ #print(paste("OverLapDNSVID:",svID))
|
|
| 1234 |
+ datOverLap<-data.frame(SVID = svID, |
|
| 1235 |
+ Probandexpression = "-", |
|
| 1236 |
+ Fatherexpression = "-", |
|
| 1237 |
+ Motherexpression = "-") |
|
| 1238 |
+ } |
|
| 1239 |
+ dataOverLap[kk,]<-c(as.character(datOverLap$SVID), |
|
| 1240 |
+ Proband_OverlapGeneExpression_TPM = as.character(datOverLap$Probandexpression), |
|
| 1241 |
+ Father_OverlapGeneExpression_TPM = as.character(datOverLap$Fatherexpression), |
|
| 1242 |
+ Mother_OverlapGeneExpression_TPM = as.character(datOverLap$Motherexpression)) |
|
| 1243 |
+ } |
|
| 1244 |
+ |
|
| 1245 |
+ ##Extracting NonOverlapped Genes |
|
| 1246 |
+ nearestUPGenes<-smapdata$Upstream_nonOverlapGenes_dist_kb |
|
| 1247 |
+ datanonOverLapUP<-data.frame(matrix(nrow=nrow(smapdata),ncol=4)) |
|
| 1248 |
+ names(datanonOverLapUP)<-c("SVID","NonOverlapUPProbandTPM",
|
|
| 1249 |
+ "NonOverlapUPFatherTPM","NonOverlapUPMotherTPM") |
|
| 1250 |
+ print("###NonOverlapUPStreamGenes###")
|
|
| 1251 |
+ for(ll in 1:length(nearestUPGenes)){
|
|
| 1252 |
+ #for(ll in 1:10){
|
|
| 1253 |
+ |
|
| 1254 |
+ datNonOverLapUP<-data.frame() |
|
| 1255 |
+ #print(paste("llUP:",ll,sep=""))
|
|
| 1256 |
+ svID<-as.character(SVID[ll]) |
|
| 1257 |
+ if(length(grep(";",nearestUPGenes[ll]))>=1){
|
|
| 1258 |
+ st1<-strsplit(as.character(nearestUPGenes[ll]),split=";") |
|
| 1259 |
+ sttemp<-as.character(st1[[1]]) |
|
| 1260 |
+ #print("1")
|
|
| 1261 |
+ gns_nonoverlap_up<-c() |
|
| 1262 |
+ for (mm in 1:length(sttemp)){
|
|
| 1263 |
+ gn_temp<-strsplit(sttemp[mm],split="\\(")
|
|
| 1264 |
+ gns_nonoverlap_up<-c(gns_nonoverlap_up, |
|
| 1265 |
+ as.character(gn_temp[[1]][1])) |
|
| 1266 |
+ } |
|
| 1267 |
+ |
|
| 1268 |
+ |
|
| 1269 |
+ datNonOverLapUP<-nonOverlapRNAseq( |
|
| 1270 |
+ gnsNonOverlap=as.character(gns_nonoverlap_up), |
|
| 1271 |
+ SVID=svID,RNASeqData=RNASeqData, |
|
| 1272 |
+ pattern_Proband=pattern_Proband, |
|
| 1273 |
+ pattern_Mother=pattern_Mother, |
|
| 1274 |
+ pattern_Father=pattern_Father) |
|
| 1275 |
+ |
|
| 1276 |
+ } |
|
| 1277 |
+ else if (length(grep("\\(",as.character(nearestUPGenes[ll])))>=1){
|
|
| 1278 |
+ #print("2")
|
|
| 1279 |
+ gnsNonOverlapUP<-strsplit(as.character(nearestUPGenes[ll]),split="\\(")[[1]][1]
|
|
| 1280 |
+ datNonOverLapUP<-nonOverlapRNAseq( |
|
| 1281 |
+ gnsNonOverlap = as.character(gnsNonOverlapUP), |
|
| 1282 |
+ SVID = svID, |
|
| 1283 |
+ RNASeqData = RNASeqData, |
|
| 1284 |
+ pattern_Proband = pattern_Proband, |
|
| 1285 |
+ pattern_Mother = pattern_Mother, |
|
| 1286 |
+ pattern_Father = pattern_Father) |
|
| 1287 |
+ |
|
| 1288 |
+ } |
|
| 1289 |
+ else{
|
|
| 1290 |
+ #print(paste("NonOverLapUPSVID:",svID))
|
|
| 1291 |
+ datNonOverLapUP<-data.frame(SVID=svID,ProbandTPM="-",FatherTPM="-",MotherTPM="-") |
|
| 1292 |
+ } |
|
| 1293 |
+ datanonOverLapUP[ll,]<-c( |
|
| 1294 |
+ as.character(datNonOverLapUP$SVID), |
|
| 1295 |
+ Proband_Upstream_nonOverlapGeneExpression_TPM = as.character(datNonOverLapUP$ProbandTPM), |
|
| 1296 |
+ Father_Upstream_nonOverlapGeneExpression_TPM=as.character(datNonOverLapUP$FatherTPM), |
|
| 1297 |
+ Mother_Upstream_nonOverlapGeneExpression_TPM=as.character(datNonOverLapUP$MotherTPM)) |
|
| 1298 |
+ } |
|
| 1299 |
+ |
|
| 1300 |
+ ##Extracting NonOverlapped Down Stream Genes |
|
| 1301 |
+ nearestDNGenes<-smapdata$Downstream_nonOverlapGenes_dist_kb |
|
| 1302 |
+ datanonOverLapDN<-data.frame(matrix(nrow=nrow(smapdata),ncol=4)) |
|
| 1303 |
+ names(datanonOverLapDN)<-c("SVID","NonOverlapDNProbandTPM",
|
|
| 1304 |
+ "NonOverlapDNFatherTPM","NonOverlapDNMotherTPM") |
|
| 1305 |
+ print("###NonOverlapDNStreamGenes###")
|
|
| 1306 |
+ for(nn in 1:length(nearestDNGenes)){
|
|
| 1307 |
+ #for(nn in 1:10){
|
|
| 1308 |
+ datNonOverLapDN<-data.frame() |
|
| 1309 |
+ # print(paste("llDN:",ll,sep=""))
|
|
| 1310 |
+ svID<-as.character(SVID[nn]) |
|
| 1311 |
+ if(length(grep(";",nearestDNGenes[nn]))>=1){
|
|
| 1312 |
+ st1<-strsplit(as.character(nearestDNGenes[nn]),split=";") |
|
| 1313 |
+ sttemp<-as.character(st1[[1]]) |
|
| 1314 |
+ #print("1")
|
|
| 1315 |
+ gns_nonoverlap_dn<-c() |
|
| 1316 |
+ for (mm in 1:length(sttemp)){
|
|
| 1317 |
+ gn_temp<-strsplit(sttemp[mm],split="\\(")
|
|
| 1318 |
+ gns_nonoverlap_dn<-c(gns_nonoverlap_dn,as.character(gn_temp[[1]][1])) |
|
| 1319 |
+ } |
|
| 1320 |
+ datNonOverLapDN<-nonOverlapRNAseq(gnsNonOverlap=as.character(gns_nonoverlap_dn), |
|
| 1321 |
+ SVID = svID,RNASeqData = RNASeqData, |
|
| 1322 |
+ pattern_Proband = pattern_Proband, |
|
| 1323 |
+ pattern_Mother = pattern_Mother, |
|
| 1324 |
+ pattern_Father=pattern_Father) |
|
| 1325 |
+ |
|
| 1326 |
+ } |
|
| 1327 |
+ else if (length(grep("\\(",as.character(nearestDNGenes[nn])))>=1){
|
|
| 1328 |
+ # print("2")
|
|
| 1329 |
+ gnsNonOverlapDN<-strsplit(as.character(nearestDNGenes[nn]),split="\\(")[[1]][1]
|
|
| 1330 |
+ datNonOverLapDN<-nonOverlapRNAseq( |
|
| 1331 |
+ gnsNonOverlap = as.character(gnsNonOverlapDN), |
|
| 1332 |
+ SVID = svID, |
|
| 1333 |
+ RNASeqData = RNASeqData, |
|
| 1334 |
+ pattern_Proband = pattern_Proband, |
|
| 1335 |
+ pattern_Mother = pattern_Mother, |
|
| 1336 |
+ pattern_Father = pattern_Father) |
|
| 1337 |
+ |
|
| 1338 |
+ } |
|
| 1339 |
+ else{
|
|
| 1340 |
+ #print(paste("NonOverLapDNSVID:",svID))
|
|
| 1341 |
+ #print ("SVID")
|
|
| 1342 |
+ datNonOverLapDN<-data.frame(SVID=svID, |
|
| 1343 |
+ ProbandTPM = "-", |
|
| 1344 |
+ FatherTPM="-",MotherTPM="-") |
|
| 1345 |
+ } |
|
| 1346 |
+ datanonOverLapDN[nn,]<-c(as.character(datNonOverLapDN$SVID), |
|
| 1347 |
+ NonOverlapDNprobandEXP=as.character(datNonOverLapDN$ProbandTPM), |
|
| 1348 |
+ NonOverlapDNfatherEXP=as.character(datNonOverLapDN$FatherTPM), |
|
| 1349 |
+ NonOverlapDNmotherEXP=as.character(datNonOverLapDN$MotherTPM)) |
|
| 1350 |
+ } |
|
| 1351 |
+ |
|
| 1352 |
+ dataFinal<-data.frame(smapdata,dataOverLap[,2:ncol(dataOverLap)], |
|
| 1353 |
+ datanonOverLapUP[,2:ncol(datanonOverLapUP)], |
|
| 1354 |
+ datanonOverLapDN[,2:ncol(datanonOverLapDN)]) |
|
| 1355 |
+return(dataFinal) |
|
| 1356 |
+ |
|
| 1357 |
+} |
|
| 1358 |
+ |
|
| 1359 |
+ |
|
| 1360 |
+ |
|
| 1361 |
+ |
|
| 1362 |
+ |
|
| 1363 |
+ |
|
| 1364 |
+ |
|
| 1365 |
+ |