Enrichment analysis based on genetic location

[LouiseT]

Hi everyone
I am integrating data analysis of data from a number of methods. Therefore my identified variants have different forms: CNVs, point mutations, indels, CpGs etc.

This, together with the fact that I wish to analyze variants located outside coding genes, complicated the use of available software.

Does anyone know a tool which can identify enrichment based on genetic location? E.g. identify if a number of my cases have variants located to e.g. chr1:1000-2000
The tool needs to function on different types of variants – therefore it does not work for me if it is depending on SNPs or other features depending on one method.

An alternative approach is to annotate my variants and then evaluate for enrichment based on IDs – however this is complicated by the fact that I want to include varianmts associated with all functional elements both coding and noncoding, TF binding sites, Enhancers etc.
Does anyone know of a tool that can handle this challenge?

[AlexReynolds]

If you have your annotations in a set of BED files, you could use the BEDOPS suite tool bedops to retrieve elements overlapping specified ranges, e.g.:

$ echo -e "chr1\t1000\t2000" | bedops --element-of -1 annotations.bed - > annotations_overlapping_chr1:1000-2000.bed

If you have a set of regions, say, in a file called my_regions.bed, then you can specify that directly:

$ bedops --element-of -1 annotations.bed my_regions.bed > annotations_overlapping_my_regions.bed

If you want both the regions and the annotations, then use the bedmap tool, e.g.:

$ bedmap --echo --echo-map --delim '\t' my_regions.bed annotations.bed > regions_with_overlapping_annotations.bed

The bedmap tool also does statistical analysis of mapped elements, if you have score or other data in your annotations. The documentation offers more detail.