Does Bowtie2 index use haplotypes and patches? and is the reference genome of Bowtie2 from UCSC or ensembl?

Question

[salamandra]

STAR manual recommends to exclude haplotypes and patches from reference genome while keeping unplaced scaffolds when aligning RNA-seq reads. Is the same recommended for aligning ChIP-seq reads with Bowtie2? Do Bowtie2 index take that into account?

Also, does Bowtie2 work with UCSC or ensembl genome? Ask cause we shouldn't 'mix' UCSC genome and ensembl genome references and as I did RNA-seq with ensembl and want to eventually compare results with ChIP-seq...

Answer 1

[Devon Ryan]

The recommendations for STAR apply to all aligners except BWA mem and novoalign. This is also true for all types of sequencing experiments.

If you used Ensembl genome names for your RNAseq dataset then you should do the same for all other datasets you plan to compare against. In general, choose one chromosome naming system for your entire lab or institute (e.g., all analyses performed where I work use Ensembl chromosome names).

[salamandra]

So, just have to make sure chr names output on Sam files by Bowtie2 are in the same format as those of RNA-seq analysis I did, and if not, just changing the chromosome names is ok?

[Devon Ryan]

Yes

[salamandra]

Noticed now that there's a command that generates bowtie2 index from genome reference: bowtie2-build. then don't get the chromosome format problem.