How to process the raw h5ad?

[jordan841220]

I think the tutorial in Jupyter notebook is unclear that it took me a while to realize how to preprocess my raw .h5ad to a SAM object...And I actually found my answer in issue #66 .

It was clear that "When loading in raw data, SAMap will automatically process them with SAM and save the results to an *_pr.h5ad file." However, the information of how to "loading in raw data" is missing.

[Lil-Psilocybe]

Hello - I am reviving this thread since I have an issue right around this same step. If you open the jupyter notebook in something like vscode you should be able to see this part. It reads:

Instantiate the SAMAP object as below.

(Note that for the sake of this tutorial, the datasets have already been processed with SAM so we will load in the SAM objects directly.)

sm = SAMAP( filenames, f_maps = 'example_data/maps/', save_processed=True #if False, do not save the processed results to *_pr.h5ad )

What I can't figure out for the life of me is when I put in my own h5ad files here and the path to my maps and press enter on vscode, there's no output at all. Any tips here?

[Lil-Psilocybe]

update! I had to change the code block from a "raw" block to a "python" block in the jupyter notebook. This got SAM to run, although I did get an error:

/home/FCAM/jbriseno/miniconda3/envs/SAMap/lib/python3.9/site-packages/umap/umap_.py:1945: UserWarning: n_jobs value 1 overridden to 1 by setting random_state. Use no seed for parallelism.
warn(f"n_jobs value {self.n_jobs} overridden to 1 by setting random_state. Use no seed for parallelism.")

[Lil-Psilocybe]

update: have to make sure I'm using peptides/transcripts with SAME ID as the sc/sn h5ad dataset. I got
0 esgene symbols match between the datasets and the BLAST graph. 0dmgene symbols match between the datasets and the BLAST graph. \ 0mg gene symbols match between the datasets and the BLAST graph.

for my SAM objects so had to go back and make sure I got data with shared gene IDs