search

omnideconv / SimBu

Simulate pseudo-bulk RNAseq samples from scRNAseq expression data
http://omnideconv.org/SimBu/
GNU General Public License v3.0
12 stars 1 forks source link

K-mer signatures #6

Closed FFinotello closed 2 years ago

FFinotello commented 2 years ago

Hey @alex-d13,

As discussed, it would be nice to have some preliminary results supporting the rationale of having also k-mer counts (as opposed to gene counts) to derive cell-type-specific signatures.

We found some issues with the pseudoaligners, but maybe there are some other promising tools out there.

For instance, I see that this scSimClassify tool has a "k-mer matrix generation module". Maybe worth a try on the Tabula Muris data or any other SMART-seq2 dataset.

Cheers, Francesca

alex-d13 commented 2 years ago

Hi,

yes, the scSimClassify tool really does sound exactly like something we could use, but i believe their repository is in a early phase. At least I cannot find any documentation or code comments, so I could not figure out yet how to run their k-mer matrix module..

Markus mentioned this paper previously, where they compared k-mer counting tools; maybe I could use one of these recommended ones (KMC3, DSK, and Gerbil).
What do you think?

grst commented 2 years ago

closing this as this has moved to its own project.