search

Nanostring-Biostats / InSituType

An R package for performing cell typing in SMI and other single cell data
Other
29 stars 11 forks source link

Implementation of float matrices #208

Open roanvanscheppingen opened 3 months ago

roanvanscheppingen commented 3 months ago

Currently, InSituType takes only raw count matrices. This is logical, since scaling or transforming data might differ per dataset. However, the inability to take count matrices consisting out of floats instead of integers limits the use of InSituType to some extend.

Arguably, proseg is a way of improving segmentation data. https://github.com/dcjones/proseg , after proseg, an expected-count matrix is provided. This consists out of floats, but should still be considered 'raw' counts before any further normalisation. Proseg has the ability to force integer counts (--output-maxpost-counts), however it might also be useful to be able to take float count matrices into InSituType processing

github-actions[bot] commented 3 months ago

Thank you for contacting us about our tools! To receive assistance, kindly email support@nanostring.com with detailed information about your issue. If applicable, attach a screenshot of any encountered errors and include a copy of the modified script in Notepad. Our customer support team will help facilitate a review and resolution of the issue.

Thank you for choosing NanoString, NanoString Dev Team

patrickjdanaher commented 3 months ago

Great point. Because insitutype for RNA uses a negative binomial model, it's stuck with using integers. That said, the new Protein mode - coming in days - runs with a Gaussian model. It could easily be run from proseg outputs.