Hello,
Thank you for releasing IsoQuant 4.0.0. I noticed that the new version includes substantially improved accuracy for poly(A)/TES and TSS positions and can incorporate these terminal sites into transcript discovery.
I am analyzing bulk Kinnex RNA-seq data from 30 tissues. My original IsoQuant 3.9 command for each sample was:
~/01.Biosoft/03.Sif/isoquantv3.9.simg \
/opt2/IsoQuant-3.9.0/isoquant.py \
-d pacbio_ccs \
--bam "$BAM" \
--reference "$ref" \
--genedb "$gtf" \
--complete_genedb \
--output "$ID.IsoQuant" \
--prefix "$ID.IsoQuant" \
-t 50
I have a few questions about migrating this analysis to IsoQuant 4.0.0:
-
Are the improved poly(A)/TES and TSS detection methods enabled automatically when running reference-based transcript discovery, or do I need to specify additional options such as --novel_apa, or other parameters?
-
I understand that TSS prediction may require --fl_data. My current input consists of all CCS reads rather than deduplicated transcript sequences. Since CCS indicates read accuracy but does not necessarily guarantee that both transcript ends are complete, what preprocessing steps or QC criteria would you recommend before treating these reads as full-length and enabling --fl_data?
Thank you very much for your help.
Best,
Dreames
Hello,
Thank you for releasing IsoQuant 4.0.0. I noticed that the new version includes substantially improved accuracy for poly(A)/TES and TSS positions and can incorporate these terminal sites into transcript discovery.
I am analyzing bulk Kinnex RNA-seq data from 30 tissues. My original IsoQuant 3.9 command for each sample was:
I have a few questions about migrating this analysis to IsoQuant 4.0.0:
Are the improved poly(A)/TES and TSS detection methods enabled automatically when running reference-based transcript discovery, or do I need to specify additional options such as --novel_apa, or other parameters?
I understand that TSS prediction may require --fl_data. My current input consists of all CCS reads rather than deduplicated transcript sequences. Since CCS indicates read accuracy but does not necessarily guarantee that both transcript ends are complete, what preprocessing steps or QC criteria would you recommend before treating these reads as full-length and enabling --fl_data?
Thank you very much for your help.
Best,
Dreames