Closed fayerodgers closed 5 years ago
@fayerodgers I may have fixed this already (unsure). Could you retry against master (should be 2.4.0-snapshot)?
If not, would it be possible to provide BAM data and/or a JBrowse directory at least with the relevant read track and sequences to speed up the fix?
Thanks,
Cheers, will retry with master and let you know!
So it looks like with the current master I can't make any genes from RNAseq reads at all (though making pseudogenes and ncRNAs from them seems to work). I've copied an example jbrowse directory here for you: ftp://ftp.ebi.ac.uk/pub/databases/wormbase/STAFF/fr7/t_muris_jbrowse/ (both of the BAMs in the BAM directory should illustrate the problem). Cheers
@fayerodgers Okay. I will take a look.
The problem is that https://github.com/GMOD/Apollo/blob/master/client/apollo/js/JSONUtils.js#L276 fails because the test_feature has no type.
https://github.com/GMOD/Apollo/blob/master/client/apollo/js/JSONUtils.js#L215 needs to be called from DraggableHTMLFeatures
@fayerodgers
You should be able to use master to add genes now.
Can you give me the location and what you expect to see (if it is not working the way you want still).
We changed some things to fix this behavior in #1788. As such, we created an exon for each match when you drag it up (as opposed to a single large exon).
Sorry @fayerodgers I’m being dense. I’ll see if I can’t come up with a fix that preserves the exon over the indels.
@fayerodgers this should be working on master now. Let me know if you run into any problems with it (or expect different behavior).
That's fixed it, thanks very much for the speedy response!
When making a gene model from a full length RNASeq read, indels seem to be being interpreted as introns. Eg, when making annotations from the top two reads here:
This is on the master branch from 19th Feb (2.3.1-snapshot).
Thanks!