bli25 / RSEM_tutorial

A short tutorial on how to use RSEM
GNU General Public License v3.0
133 stars 46 forks source link

RSEM low alignment rate on bowtie2 built index from full mouse genome file with gtf file provided #1

Open Jiglov opened 6 years ago

Jiglov commented 6 years ago

Hi, there,

I've ran RSEM on my data and saw alignment rate of around 2% Example: "3134964 reads; of these: 3134964 (100.00%) were paired; of these: 3054835 (97.44%) aligned concordantly 0 times 52834 (1.69%) aligned concordantly exactly 1 time 27295 (0.87%) aligned concordantly >1 times 2.56% overall alignment rate"

I've aligned this data using hisat2 with the same full mouse genome file and the alignment rate was >90%, doing it on Galaxy with Hisat also resulted with >90% alignment rate.

I used bowtie2 in RSEM command to generate alignment based on the index built by bowtie2 also using RSEM:

This was used to build the index: rsem-prepare-reference --gtf $genes --bowtie2 ~/RNAseq/RSEM/Mus_musculus.GRCm38.dna.chromosome.1.fa ~/RNAseq/ref/mouse_ensemble

And this was used to run RSEM within the for loop: rsem-calculate-expression -p 8 --bowtie2 --paired-end --strandedness forward --append-names --estimate-rspd --output-genome-bam ~/RNAseq/RSEM/c_all/Control"$i"_R1.fastq.gz ~/RNAseq/RSEM/c_all/CTRL"$i"_R2.fastq.gz ~/RNAseq/ref/mouse_ensemble ~/RNAseq/RSEM/c_all/RSEM/Control"$i"

I've looked online to check this issue found a google group discussion. https://groups.google.com/forum/#!topic/rsem-users/3RiF4WneNw0

Did not find it helpful.

I'm trying to understand if that what it is suppose to be in the report, there are no separate files with unaligned reads....

ArnonMe commented 4 years ago

same problem.