I am trying to perform exon-intron split analysis on human RNASeq datasets, using eisaR. Since I am not able to generate the genome index files using my computer, I downloaded the index files from Hisat2 website. I would like to know if it would be possible to use these files for further alignment step, instead of generating the index files afresh on my computer. If this is possible, what changes need to be made in the command lines? Would appreciate any guidance on this since I am new to the field.
If you are creating the alignments using qAlign(..., aligner = "Rhisat2"), it is recommended to start from reads and genome sequence and let QuasR perform both genome indexing and alignments, for example using the BSgenome and TxDb packages available from Bioconductor. This will ensure that the genome and annotation are consistent, and that alignments have been created as expected by eisaR. Furthermore, index creation will only have to be performed once and typically will not take more time than aligning the reads.
It is possible to align your reads outside of QuasR (not using qAlign), by performing the following steps:
download genome, genome-index and matching annotation
use this annotation to create regions of interest with eisaR
download and install hisat2 (or any other aligner)
align your reads outside of R using parameters that ensure up to a single alignment per read
if necessary, convert resulting alignments to bam format and index bam files
list bam file names in a "samples text file" as described in the QuasR vignette
run qAlign(...) on that "samples text file" (no new alignments will be created, the existing bam files will be used)
use the returned qProject object to quantify exonic and intronic regions using qCount(...) and eisaR
Please note that in this case, it will not be possible to enforce a maximal number of alignments per read (maxHits parameter from qAlign), as hisat2 does not support this concept and QuasR has no way to enforce it for bam files that are created outside of R.
Dear Dr. Stadler,
Thank you for your guidance. I will follow your suggestions.
Thank you
Yours sincerely Nishanth