Hi the community!

I've recently written a toolkit which allows to annotate scRNA-seq data directly from literature. It requires a raw adata file and a pdf file. Then It can automatically extract information from articles and do all downstream job like filter, preprocessing, clustering, annotation. It now works smoothly, and I want to get some advices to this tiny project.

I think the target users may be:

1. Need to add a public resource to their own dataset, but find only raw file in GSE
2. Want to do Data Integration to a large datasets, using labeled data may out-perform label-free data
3. Want to automatically create a super-large single cell dataset (like all the pubilc dataset, maybe a little crazy)