hisat2 lead to highly multimapping rate
I'm doing RNAseq analysis, and I did mapping with hasit2. It give me a high total mapping rate(94%), but also high multimapping rate(20%). I use hisat2 mapping another RNAseq data which has been tophat aligned(total mapping 81.66%,unique mapping (78.95%), similarly, I get high total mapping rate(95.29%), but also high multimapping rate(38%). It make me confused. I build index with genome and transcriptome , I'm wondering if this cause this ouput? someone can give me a hint?
• 4,674 views
•
link
written
by
yuxinghai •
0 answers
No answers yet.
Log in to answer this question.
More posts like this
-
High amount of intronic/intergenic reads in SMARTer stranded total bulk RNAseq
written by Mat •
-
BAM file contains much more reads than expected
written by brian •
-
Salmon lib_format_counts.json - explanation
written by Lada •
-
featureCounts low rate and strange output
written by Yi •
-
featureCounts: extremely low rate of 'Successfully assigned alignments'
written by Yi Chu •
-
Best method for align EST data to genome
written by nisaji27 •
-
featureCounts for paired bam files and dealing with multimapping reads?
written by bioinfo8
-
tophat did not align all input reads
written by elainzp •
-
Mouse RNASeq, 50% discordant reads
written by jc.szamosi •
-
Low final mapping rate by TopHat in spite of high initial Bowtie mapping
written by drtamermansour •
Which parameters are you using in the mapping? how long are your reads? does the quality control look fine?
hisat2 -p 8 --dta --no-mixed --no-discordant -x /data3/zhoulab/yuxinghai/anotiation/grch38_tran/genome_tran --un G1/G1.tran_unmap.sam -1 G1/G1_R1.fq -2 G1/G1_R2.fq -S G1/G1.tran_map.sam
read length? does your QC look file?
PE 150, and I did fasqc, the base quality all almost >28
G+C, overrepresented sequences?
through fastqc, I can see two peaks,a little differnent with other RNAseq. but It's just a normal RNAseq. sorry, I don't know how to upload the picture.
Sorry for my late answer. I have been away. I would find suspicious to see two peaks in the G+C distribution. Have you blast the overerepresented sequences?
I am also wondering why so high multiple alignment rate by HISAT2. I am using "NH:i:1" tag to extract uniquely mapped read, but this kind of unique reads seems much more than that number told by HISAT2.