create data for human browser on demo server

[nathandunn]

Variants:

• [ ] links rsABCD to cvicdb and GA4GH / UCSC, dbSNP
• [ ] make vertical lines into dots / with varying sizes / colors

Update Evidence tracks:

• [ ] implement using updated gene models (see #481) for evidence tracks
• [ ] implement using updated annotation data (also #481)

Relevant data types . . add to WA2:

• [x] add genbank data (draggable)
  • [x] recollect raw data
  • [x] fix chr1 configuration
  • [x] default to Canvas track (for now)
  • [x] chr1 is misconfigured
  • [x] fix gene look and feel for draggable
• [x] verify we can see BRCA1 and BRCA2 variant and gene data
• [x] verify link: http://icebox.lbl.gov/Apollo2/jbrowse/index.html?overview=1&nav=1&loc=chr1%3A97230113..97691114&tracks=Annotations%2CHomo_sapiens.GRCh38.81.gff3&tracklist=1&highlight=&organism=5402342
• [x] add human refeqs
  • [x] refetch raw data (working)
  • [x] map via reference: http://www.ncbi.nlm.nih.gov/projects/genome/assembly/grc/human/data/
  • [x] reprocess (it looks good)
• [ ] clinvar:
  • [ ] get data
  • [ ] add track
• [ ] dbsnps: http://www.ncbi.nlm.nih.gov/projects/SNP/snp_summary.cgi
  • [ ] get data
  • [ ] add track
• [ ] 1K genomes
  • [ ] get data (working)
  • [ ] configure tracks
  • [ ] add track
• [x] rerun generate_names.pl

[nathandunn]

JBrowse for reference: http://jbrowse.org/ucsc/hg19/touch.html?loc=chr1:208,056,300..208,073,080&tracks=DNA,knownGene,ccdsGene,snp131,pgWatson,simpleRepeat

. . I think that we'll be using hg38 unless there is a reason not to.

[selewis]

Check with Nicole. I think there may have been an update. Important to have the up to date version.

On Thu, Jul 23, 2015 at 8:52 PM, Nathan Dunn notifications@github.com wrote:

JBrowse for reference: http://jbrowse.org/ucsc/hg19/touch.html?loc=chr1:208,056,300..208,073,080&tracks=DNA,knownGene,ccdsGene,snp131,pgWatson,simpleRepeat

. . I think that we'll be using hg38 unless there is a reason not to.

— Reply to this email directly or view it on GitHub https://github.com/GMOD/Apollo/issues/499#issuecomment-124315331.

[nathandunn]

@nlwashington let me know how if this sounds like a good starting point (the checklist above). I expect we'll do a bit of iteration over it.

http://icebox.lbl.gov/WebApolloHuman/jbrowse/?loc=chr11%3A476355..654095&tracks=DNA%2CAnnotations%2COGS-GRCh38%2COGS-GRCh38-gene&highlight=

[cmdcolin]

I mentioned this awhile back but the gene track is getting confused by the presence of some of the pseudogenes for some reason (I don't have the data file but generally I would load pseudogenes as a separate track anyways). I would load just the genes from the file using either using the --type "gene" argument for a CanvasFeatures type track or --type "mRNA" in HTMLFeatures. We would normally prefer HTMLFeatures in WebApollo anyways so --type "mRNA" is most likely the best option

[cmdcolin]

The Red highlighted features in the above image are technically a glitchy pseudogene, probably just need to clean up the GFF if these need to be loaded cc @deepakunni3

[nathandunn]

definitely need to clean up the GFF . . . I don't even think that all of the chromosomes load.

[nathandunn]

@monicacecilia @selewis This is the human server corresponding to our deliverables and what we'd put in the grant (as opposed to icebox):

https://genome.monarchinitiative.org/apollo/507/jbrowse/index.html?loc=chr7:102240331..102240644&tracks=dbsnps,clinvar,GenBank%20TopLevel%20MRNA

I'm not sure which of this we want to move over. The ClinVar and dbSnp data is manageable, but the modEncode data is massive.

[nathandunn]

Kind of done