New mouse reference genome (GRCm39/mm39)

Question

[sarahmanderni]

Hi,

I am planning to perform sequencing alignment using the new mouse reference genome mm39 available on UCSC: https://hgdownload.soe.ucsc.edu/goldenPath/mm39/bigZips/ (bowtie2 on chipseq data). Did I understand correctly that both mm39.2bit and mm39.fa.gz are the same and I can use either of them (2bit file first convert to fasta) for the alignment? And one practical question: have you noticed significant overall difference compared to alignment against mm10?

[biofalconch]

Did I understand correctly that both mm39.2bit and mm39.fa.gz are the same and I can use either of them (2bit file first convert to fasta) for the alignment?

Yes, 2bit is a format that stores DNA sequences in a compact manner (I believe it even includes masked regions), but I wouldn't go through the hassle of converting it to fasta format if there is already a fasta file.

Answer 1

[Pierre Lindenbaum]

Did I understand correctly that both mm39.2bit and mm39.fa.gz are the same

fa.gz is a fasta text file compressed with gzip

.2bit is a UCSC specific format, where each nucleotide is packed in bytes, mostly used by blat

bowtie2 requires you to build a specific index from the fasta.

[sarahmanderni]

Thanks for the response. So the content for mm39.2bit and mm39.fa.gz are the same regardless of the format and I can decide which one to use. I just wanted to make sure as UCSC has used "Soft-masked" term for mm39.fa.gz file in the readme.txt file but not for the mm39.2bit (they used "complete" reference for mm39.2bit).