Entering edit mode
3.6 years ago
z.ziriat
•
0
Hello everyone, I have a problem when using FeatureCounts to assign my Nanopore. When using it with a Bam file i always have 0 total alignment recognized, and so i have no successfully assigned alignment. I think the problem is that FeatureCounts refuse to read my bam, or cannot read it for some reason (when i use it for another Bam resulting from another Run i have no problem with it).
Can you help me please ?
Ps : I've aligned with minimap2, sorted and indexed, and I've also checked the gtf chromosome format.
Have you tried this? RNA-SEQ data, reads count zero
Yes i tried it, but the problem doesn't come from the assignment, but from the number of reads Feature count recognize. and it recognize nothing
Since you have long read data it is possible that your reads are overlapping multiple features (e.g. exons) in your reference. You may need to use
-O(overlap) option to get counts. I suggest you take a look atfeatureCountsmanual.I will add to this that featureCounts is a tool designed for short-read counting.
The entire process may make a number of assumptions that may have made sense for short reads but may not hold for long reads. For example what if the alignment of the long read bridges over a very large region (with a large deletion). Does the read overlap with the deleted region or not? What does featureCounts do in such cases? It is not so obvious.
You should really think about whether long reads will work as desired.