cxy
Dear MAJIQ Team,
I hope this message finds you well. I have a few questions regarding the usage of the voila lr command in MAJIQ-L:
1. Is it possible to use the output from StringTie (i.e., .gtf and corresponding expression files) as input for:
voila lr \
--lr-gtf-file /PATH/TO/SAMPLE_ID.gtf \
--lr-tsv-file /PATH/TO/SAMPLE_ID.t_data.ctab \
-sg /PATH/TO/splicegraph.sql \
-o OUTPUT_FOLDER
Since StringTie does not provide raw read counts by default, can TPM or coverage (cov) values from t_data.ctab be used in place of counts in the .tsv file?
2. Does voila lr only support visualization per individual sample, or is it also possible to use a merged GTF (e.g., from multiple samples) along with a combined expression .tsv for a more global overview?
3. Lastly, I’m very interested in the sQTL analysis functionality. May I ask if there is an estimated timeline for when this feature will be available in MAJIQ or VOILA?
Thank you very much for your time and for your continued development of such a valuable toolkit for alternative splicing analysis.
Best regards,
Xinyi Chang
seonghan
Hi Xinyi,
Let me answer #1,2 together and San can answer #3.
MAJIQ-L accepts transcript read number, not TPM, to compute long read junction reads. Also, yes, VOILA LR supports visualization per individual sample. Technically you can give one merged GTF + merged transcript read as long it matches with the short reads.
Best,
Seong
seonghan
Hi Xinyi,
Unfortunately, the current VOILA LR doesn’t support deltapsi.voila, so you need to compare/contrast SR & LR using psi.voila. I recommend comparing VOILA LR for control vs disease separately using psi.voila. Regarding grouping or comparing long-read results, you can merge control samples and disease samples separately to have two gtf files, but for details, I recommend asking the IsoQuant team.
Best,
Seong