htslib VS MMseqs2

There is another nice multi-core gzip based library called BGZF[1]. It is commonly used in bioinformatics. BGZF has the added advantage that it is block compressed with built in indexing method to permit seeking in compressed files.

[1] https://github.com/samtools/htslib

In contrast, processing can start immediately and only transfer what is necessary if tooling can read bytes of data directly from Amazon S3. Tools based on htslib can do this, so you can run something like:

Another idea is add in reading cloud data natively e.g. htslib which samtools relies on can read s3 directly. https://github.com/samtools/htslib

Greetings! I'm a bioinformatics software dev in San Francisco and I'm looking for others interested in working on a high-performance genetic data analysis project. The project is in C, using https://github.com/samtools/htslib.

I'm working on a bioinformatics-related binding called ruby-htslib. htslib makes heavy use of bit fields throughout the library, so supporting bit fields is inevitable.

A tool I often recommend for sequence clustering is mmseqs2 : https://github.com/soedinglab/MMseqs2, fast and efficient :)

Tools like the the mmseqs2 "taxonomy" module, or DIAMOND v2, can efficiently align contigs to genome databases to assign taxonomy, but it seems like they aren't intended to provide abundance estimates for each taxon (since that would require mapping reads, and mmseqs2 can't even use paired-reads). Can anyone recommend tools or methods for A) connecting per-contig coverage information to contig taxonomy, or B) mapping short reads against genome databases?