Region white-matter enrichment trends
Closed this issue · 7 comments
From the snRNA-seq data so far, we've seen that the post region samples (posterior) seem to have higher proportions of oligos and other white-matter enriched genes. Note that this is a subset (n = 19) of the 30 spatial samples we have. But with those n = 19, we could make a scatterplot comparing the proportion of WM spots observed in the spatial data (screenshot incorrectly says GM) vs the proportion of oligos (and similarly for other cell types) in the snRNA-seq data. Then color the samples by region (ant, mid, post). It could be that we have 3 different lines, or maybe all 3 lines have the same slope but different intercepts.
To extract the proportion of WM, we can use the k = 2 BayesSpace results from https://github.com/LieberInstitute/spatialDLPFC/blob/main/plots/03_BayesSpace/polychrome_vis_grid_clus_sfigu_BayesSpace_k2.pdf where cluster 1 was WM if I remember correctly. The SPE at /dcs04/lieber/lcolladotor/spatialDLPFC_LIBD4035/spatialDLPFC/processed-data/rdata/spe/01_build_spe/spe_filtered_final_with_clusters.Rdata likely has this info already. If not, you can use http://research.libd.org/spatialLIBD/reference/cluster_import.html (Nick has used it).
@lahuuki we haven't checked this yet, right?
I know this is the least important open issue we have, but well, at some point, I'd like to see these plots. This could be done after our bioRxiv submission though.
without SP2806, 16, 17, 20 and 28 instead of Sp02D01
from #140 (comment)
The current plot at https://github.com/LieberInstitute/spatialDLPFC/blob/main/plots/spot_deconvo/05-shared_utilities/prop_WM_position_scatter.pdf is with Ambiguous being part of the %. Can you make it again without it? (Have the 2 versions)
Do you have updates on this one @Nick-Eagles?
I made the new version, and the oligos look a little better (more correlated, as expected)
Thanks Nick. We discussed this plot today in the deconvolution meeting and will likely use a version of this (with the RNAscope / HALO data too) for the deconvolution commentary paper.