[BioC] VariantAnnotation: fine define Locating variants in and around genes

[Valerie Obenchain]

Hi Fabrice,

To identify snps (or any ranges) in introns only, use IntronVariants() 
as the 'region' argument. The CodingVariants are the exon regions. If 
you want all regions except coding, I would suggest using AllVariants().

This output is from the man page example. The 'loc_coding' name is 
misleading since AllVariants were use as 'region'. I have changed it to 
'loc_all' in the devel branch.

 > loc_coding <- locateVariants(vcf_adj, txdb, AllVariants())
 > loc_coding
GRanges with 16 ranges and 7 metadata columns:
              seqnames               ranges strand |   LOCATION   QUERYID
                 <Rle>            <IRanges>  <Rle> |   <factor> <integer>
                  chr1 [   13220,    13220]      * |     intron         1
                  chr1 [   13220,    13220]      * | spliceSite         1
                  chr1 [   13220,    13220]      * |     intron         1
                  chr1 [   13220,    13220]      * |     intron         1
                  chr1 [   13220,    13220]      * | spliceSite         1
...
...

This example has variants in splice sites, introns, coding and 
intergenic regions.

 > tbl <- table(loc_coding$LOCATION)
 > tbl[tbl > 0]

spliceSite     intron     coding intergenic
          2          7          2          5

The result can be subset on LOCATION for the region of interest. The 
QUERYID column maps back to the row number in the original 'query' 
argument to locateVariants().

introns <- loc_coding[loc_coding$LOCATION == "intron", ]
 > head(introns, 3)
GRanges with 3 ranges and 7 metadata columns:
    seqnames         ranges strand | LOCATION   QUERYID      TXID
       <Rle>      <IRanges>  <Rle> | <factor> <integer> <integer>
        chr1 [13220, 13220]      * |   intron         1         1
        chr1 [13220, 13220]      * |   intron         1         2
        chr1 [13220, 13220]      * |   intron         1         3


Valerie


On 01/31/2013 12:34 PM, Fabrice Tourre wrote:
> Dear list,
>
> I am using VariantAnnotation to Locate variants in and around genes.
>
> In VariantAnnotation, the region is defined as: Coding Variants,
> IntronVariants, FiveUTRVariants, ThreeUTRVariants, IntergenicVariants,
> SpliceSiteVariants or PromoterVariants.
>
> If it possible to know whether a snp is in exon/intron within
> transcription region but outside coding region?
>
> Thanks.
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>