Unable to search by gene names in a custom session with custom .genome file

[Heather Raimer Young]

Hello, IGV team!

I am using IGV 2.5.0 on a computer running ubuntu 20.04.1 LTS. 

I used the "create .genome file" feature to create a genome for the new chm13 whole human genome from the T2T consortium. I know that the .genome file has been deprecated to move towards .json files, but at this time with how new the chm13 genome is I couldn't easily see how to work with the .json files and therefore used the older .genome.

I used a FASTA file that was combined from the v1.0 draft of that genome on their GitHub (https://github.com/marbl/CHM13/blob/master/README.md)and added the Y chromosome FASTA from this NCBI link (https://www.ncbi.nlm.nih.gov/nuccore/CP086569.2) to create a custom v1.0plusY FASTA file (because the index for aligning my data with bowtie2 is based on v1.0plusY as indicated on the botwie2 index website = http://bowtie-bio.sourceforge.net/bowtie2/manual.shtml). I also used the v1.0 gff3 file from the CHM13 GitHub page for the gene annotation file.

All this to say I made a .genome file, seemingly, successfully and was able to load it and start to look at my aligned data. However, I am unable to search by gene names to quickly go to that region (as I can do with the preloaded Human hg38 genome) even though I can see the genes in the gene track in the viewer. How can I gain the ability to search by the gene names in this custom genome file? 

I am unable to attach the custom .genome file as it indicates it is too large, but would be happy to send it if there is another method. But I have attached the portion of my IGV log corresponding to when I experienced this problem. 

[igv-help]

Hi,

While it's not available in the regular genomes menu in IGV, we actually have a genome.json file for the chm13 genome.

To load the genome, select  Genomes > Load Genomes from URL and enter https://s3.amazonaws.com/igv.org.genomes/t2t-chm13-v1.1/t2t-chm13-v1.1.json

To see what the .json file looks like, you can just enter https://s3.amazonaws.com/igv.org.genomes/t2t-chm13-v1.1/t2t-chm13-v1.1.json into your web browser.

BTW, you're running a very old version of IGV.

Helga

[igv-help]

I can't really answer your question because I don't recall what search options we had in IGV 2.5.0.     However with any recent version of IGV you can search by name included in non-indexed GFF files, as long as the name you are searching for is in the gff file as a column 9 property.    For IGV 2.5.0 you might need to explicitly identify the name by creating and loading a 4 column bed file, 4th column being the name.   Again I am not sure what the options were, but that would be a safe bet.

Also,  the "json" genome file alluded to above will not work in IGV 2.5, you need a recent version for that.

 

[Heather Raimer Young]

Thank you so much for this! We will definitely update IGV in the lab (didn't realize how out of date we were until looking up the version to post this question) and start using the chm13 you already generated!

[igv-help]

It's good to hear you will update your IGV. As my colleague pointed out, you won't be able to use genome .json files with the old version. 

Helga

[James Robinson]

I've prepared a searchable bed file that should work with 2.5, loading the following bed file from the "File" menu should enable searching with your current IGV

https://s3.amazonaws.com/igv.org.genomes/t2t-chm13-v1.1/CHM13.names.bed.gz

[吉翔骏]

LIKE what IGV-help said, load the json file with the currect direction of those files, then you can search with gene symbols