Question: Alignment rate differes using hg19 and hg38
0
gravatar for fate.gh
13 months ago by
fate.gh10
fate.gh10 wrote:

I have some RNA-seq fastq files.I'm using HiSat to align the fastq files. To do so, I use the built-in genome reference in Galaxy. When I align them to hg19, the alignment rate is much higher than when I align them to hg38. What causes such big difference?

Here this is an example:

* Alignment to hg19:

format    bam
database  hg19

37328991 reads; of these:
37328991 (100.00%) were unpaired; of these:
3767939 (10.09%) aligned 0 times
16947652 (45.40%) aligned exactly 1 time
16613400 (44.51%) aligned >1 times
89.91% overall alignment rate
[bam_sort_core] merging from 24

* Alignment to hg38:

format    bam
database   hg38

37328991 reads; of these:
37328991 (100.00%) were unpaired; of these:
11718927 (31.39%) aligned 0 times
15954103 (42.74%) aligned exactly 1 time
9655961 (25.87%) aligned >1 times
68.61% overall alignment rate
[bam_sort_core] merging from 24

What should I do?

Will it cause a problem if I want to obtain read counts for DE analysis using htseq?

hisat hg19 hg39 alignment rate • 470 views
ADD COMMENTlink modified 13 months ago by Jennifer Hillman Jackson23k • written 13 months ago by fate.gh10
0
gravatar for Jennifer Hillman Jackson
13 months ago by
United States
Jennifer Hillman Jackson23k wrote:

Hello,

These are the same exact inputs and parameters?

Seems odd, but possible, it depends on the read content. Hg38 is much more polished than hg19.

Perhaps run htseq_count on both and compare? It might provide a clue about where those extra unaligned hg38 reads were previously mapped to hg19. And what those sequences contain - you could isolate them and run FastQC to do some QA on them.

Jen, Galaxy team

ADD COMMENTlink written 13 months ago by Jennifer Hillman Jackson23k

Hi,

Yes, These are the same exact inputs and parameters... I ran htseq-count and compared the results, but since the gtf files are different (different ensembl versions to match hg19 and hg38), the result are not the same.

ADD REPLYlink modified 13 months ago • written 13 months ago by fate.gh10
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 104 users visited in the last hour