Dear Friends,

My following geneRange object has all chromosomal information in one GRange object. 


library(VariantAnnotation)
library(TxDb.Hsapiens.UCSC.hg19.knownGene)
txdb = TxDb.Hsapiens.UCSC.hg19.knownGene
txg = transcriptsBy(txdb,by="gene")

> geneRange <- sort(unlist(range(txg[as.character(myGenes$geneId)]))) > geneRange GRanges with 11 ranges and 0 metadata columns: seqnames                 ranges strand <Rle>              <IRanges>  <Rle> 4067     chr8 [ 56792386,  56925006]      + 26579    chr11 [ 69061622,  69156450]      + 595    chr11 [ 69455873,  69469242]      + 2017    chr11 [ 70244612,  70282690]      + 10413    chr11 [101981192, 102104154]      + 330    chr11 [102188181, 102210135]      + 329    chr11 [102217966, 102249401]      + 9965    chr11 [ 69513006,  69519106]      - 2249    chr11 [ 69587797,  69590171]      - 2248    chr11 [ 69624736,  69634192]      - 47    chr17 [ 40023179,  40075272]      - ---

I have to get SNP and genotype information for the same from 1000 genome. Below code does that. However, each time I have to create ftp url and connect to 1000 Genome resource.
thats painstaking.

i=0
for(chr in as.vector(seqnames(geneRange))){
baseUrl <- paste("ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/phase1/analysis_results/integrated_call_sets/ALL.chr",chr,".integrated_phase1_v3.20101123.snps_indels_svs.genotypes.vcf.gz",sep="")
i=i+1
print(baseUrl)
inGene <- snpsInGene(geneRange[i],baseUrl,geno="GT",info=NA)
print(inGene)
}

If I have my geneRange information as per chromosome in GRangeList form, my script becomes more efficient. because, I can retrive genotypes from each chromosomes in one go.
So, May I know a way to convert geneRange GRange object to GRangeList Object as per chromosome number?

Thanking you in anticipation.
Regards,
Prashantha

	[[alternative HTML version deleted]]