Hisat2

Question

[Pegasus]

Hi all,

I indexed my genome assembly using hisat2-build resulting in 8 ht2 files

Used hisat2 to align my (RNA-seq paired-ends) reads to the indexed genome resulting SAM file for each pair

The alignment statistics are shown below

1. I wonder if 98.21% is considered a good alignment result.
2. I tried to blast several sequences that were left out and stored in unaligned.sam file. Surprisingly, They aligned perfectly to the same reference genome using NCBI-Blast.

So, can I include unaligned .sam file or extract these unmapped sequences and use them in further RNA-seq analysis? Or I should use another alignment tool like STAR for my bacterial RNA-seq data analysis.

Thank you

Answer 1

[ATpoint]

98% is very good. I would not even bother checking what the remaining 2% are and just continue with downstream analysis. You cannot include unaligned sequences anyway since they have no coordinates. It is fine, just proceed.

[Pegasus]

Thank you ATpoint,

Actually, I am still interested in these unmapped reads because some of them successfully matched some of the genes I am interested in.

Also, I noticed that some users used the annotation file gtf.file instead of the assembly.fasta file to build an index using hisat2-buid. So, which file is more recommended to use?