Combine annotations

[Liam_Gallagher]

Hi all...could you suggest me a way to combine annotations coming from different files and related to the same subject (e.g. the same genomic coordinate) ? I was thinking about Annovar (tool to annotate nucleotide variants, indels, etc.) which produces a lot of files...and usually I use Excel to combine all of this annotations in one final document containing all the informations.
This method works, but it takes me a lot of time....
Thank you!

Another question: how can I annotate a BED file containing a list of genomic coordinates, with corrisponding Gene Symbols ? I tried using the UCSC genome browser (tables section), but I was able to annotate just with ucsc gene names, and not with normal names (like Hugo symbol names). BEDTools can do it ? thanks!

[jxchong]

Annovar has a summarize_annovar.pl script you can take a look at...

Personally, I combine the annotations using my own scripts. Excel is definitely not that way to do it!! You're just going to end up making mistakes, plus Excel can't handle more than a handful of data sets.

[Liam_Gallagher]

Yes...I know about summarize_annovar, but I was not able to get it working...because of an error. Today I tried to redownload some database, and now it works fine!
Fortunately I have just some variant to annotate (we look just at some gene...it's not a big sequencing project..) and to do it I use a macro that I've written...and do the work automatically. But to combine all the input file to process with my macro, it takes time....so i was wondering if you know other tools that do it..however thank you for the suggestions!

What about my second question (to annotate a bed file with hgnc gene symbols) ?! :-)

Annovar has a summarize_annovar.pl script you can take a look at...

Personally, I combine the annotations using my own scripts. Excel is definitely not that way to do it!! You're just going to end up making mistakes, plus Excel can't handle more than a handful of data sets.

[jxchong]

The other tool would be writing a script yourself to do it (usually Perl or Python)

http://fasta.bioch.virginia.edu/cshl/stubbs/galaxy_bed2genes.html

[AlexReynolds]

Another question: how can I annotate a BED file containing a list of genomic coordinates, with corrisponding Gene Symbols ? I tried using the UCSC genome browser (tables section), but I was able to annotate just with ucsc gene names, and not with normal names (like Hugo symbol names). BEDTools can do it ? thanks!

BEDOPS is another suite of tools for manipulating BED data.

You can use the bedmap tool to annotate genomic regions with IDs or other data from other sets (gene names, etc.).

As an example, if you have regions in a sorted file called Regions.bed and your genes in a sorted file called Genes.bed (where gene IDs are in the fourth column, per UCSC specification), the file AnnotatedRegions.bed will contain your answer:

$ bedmap --echo --echo-map-id --delim '\t' Regions.bed Genes.bed > AnnotatedRegions.bed

The only requirement is that the inputs are sorted. Use the sort-bed utility for this purpose, e.g.:

$ sort-bed UnsortedRegions.bed > SortedRegions.bed

[Liam_Gallagher]

BEDOPS is another suite of tools for manipulating BED data.

You can use the bedmap tool to annotate genomic regions with IDs or other data from other sets (gene names, etc.).

As an example, if you have regions in a sorted file called Regions.bed and your genes in a sorted file called Genes.bed (where gene IDs are in the fourth column, per UCSC specification), the file AnnotatedRegions.bed will contain your answer:

$ bedmap --echo --echo-map-id --delim '\t' Regions.bed Genes.bed > AnnotatedRegions.bed

The only requirement is that the inputs are sorted. Use the sort-bed utility for this purpose, e.g.:

$ sort-bed UnsortedRegions.bed > SortedRegions.bed

Thank you very much for your suggestions....they are very helpful!!