Extract new clinical gene sets from the literature

[kmaynard12]

• snRNAseq MDD https://pubmed.ncbi.nlm.nih.gov/32341540/
• snRNAseq ASD https://pubmed.ncbi.nlm.nih.gov/31097668/
• DOUBLE CHECK: bulk sex differences SCZ https://www.biologicalpsychiatryjournal.com/article/S0006-3223(21)01180-X/fulltext
• snRNAseq SCZ https://www.medrxiv.org/content/10.1101/2020.11.06.20225342v1.full
• snRNAseq and spatial SCZ https://www.biorxiv.org/content/10.1101/2020.11.17.386458v2
• Literature search for other new datasets that have been published since 2020 (single cell or bulk)
• Find relevant supplementary tables for differentially expressed genes between cases and controls (and by cell type for snRNAseq)
• Extract gene lists and decide a cut off value for significant genes to include for registration with spatial data @lcolladotor might be able to help with assigning appropriate cut off (might also be relevant for @shkwon17 for AD project)
• how to use a gene set when they don't provide ensembl id. maps_ids function?
• read ASD study I added and figure out they generated their gene sets. Do they have them by cluster and could we the make an entire separate heat map comparing out data to theirs.
• drop double white matter clusters and meninges cluster?? Does the enrichment use the top 100 genes and should we use more?

[lcolladotor]

@abspangler13 where where you doing this and how are along are you?

[abspangler13]

I performed it for the k= 9 data set against all of the datasets from the pilot study and two new datasets that I added. Here's the code for the two new datasets I added as well as some comments about two sets we were interested in adding.

spatialDLPFC/code/analysis/10_Clinical_Gene_Set_Enrichment/01_Clinical_Gene_Set_Enrichment.R

Lines 251 to 306 in ea6c27c

	##############
	#### Nagy sn_rna_seq in dlpfc in MDD
	#### https://www.nature.com/articles/s41593-020-0621-y
	#### sup table 6 is marker genes
	#### sup table 32 is DEGs
	#### file = /dcs04/lieber/lcolladotor/spatialDLPFC_LIBD4035/spatialDLPFC/processed-data/rdata/spe/10_clinical_gene_set_enrichment/41593_2020_621_MOESM3_ESM.xlsx
	#############
	mdd <- as.data.frame(read_excel("/dcs04/lieber/lcolladotor/spatialDLPFC_LIBD4035/spatialDLPFC/processed-data/rdata/spe/10_clinical_gene_set_enrichment/41593_2020_621_MOESM3_ESM.xlsx", sheet = "Supplementary Table 32", skip = 2, ))
	# need to somehow get gene_id or ensemblID
	ens4 <- select(org.Hs.eg.db,
	columns = c("ENSEMBL", "ENTREZID", "SYMBOL"),
	keytypes = "SYMBOL",
	keys = as.character(unique(mdd$Gene))
	)
	mdd_geneList <- with(
	mdd,
	list(
	# DE_PE_ASD.Up = ensembl_gene_id[ASD.t.value > 0 & ASD.fdr < 0.05],
	# DE_PE_ASD.Down = ensembl_gene_id[ASD.t.value < 0 & ASD.fdr < 0.05],
	# DE_PE_BD.Up = ensembl_gene_id[BD.t.value > 0 & BD.fdr < 0.05],
	# DE_PE_BD.Down = ensembl_gene_id[BD.t.value < 0 & BD.fdr < 0.05],
	# DE_PE_SCZ.Up = ensembl_gene_id[SCZ.t.value > 0 & SCZ.fdr < 0.05],
	# DE_PE_SCZ.Down = ensembl_gene_id[SCZ.t.value < 0 & SCZ.fdr < 0.05]
	)
	)
	##############
	### snRNAseq ASD
	### https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7678724/#SD5
	### data S4 is DEGS
	### data S3 is marker genes
	### file = /dcs04/lieber/lcolladotor/spatialDLPFC_LIBD4035/spatialDLPFC/processed-data/rdata/spe/10_clinical_gene_set_enrichment/NIHMS1053005-supplement-Data_S4.xls
	#############
	asd_rnaseq <- as.data.frame(read_excel("/dcs04/lieber/lcolladotor/spatialDLPFC_LIBD4035/spatialDLPFC/processed-data/rdata/spe/10_clinical_gene_set_enrichment/NIHMS1053005-supplement-Data_S4.xls", sheet = "ASD_DEGs"))
	asd_rnaseq <- clean_names(asd_rnaseq)
	asdRNA_geneList <- list(
	DE_ASD_RNA.Up = asd_rnaseq$gene_id[asd_rnaseq$fold_change > 0 & asd_rnaseq$q_value < 0.05],
	DE_ASD_RNA.Down = asd_rnaseq$gene_id[asd_rnaseq$fold_chang < 0 & asd_rnaseq$q_value < 0.05]
	)
	############
	### https://www.medrxiv.org/content/10.1101/2020.11.06.20225342v1.full
	### snRNAseq SCZ
	### supplementary table 6 is DEGS, can't figure out how to download
	##########
	##########
	### snRNAseq and spatial SCZ
	### https://www.biorxiv.org/content/10.1101/2020.11.17.386458v2
	### supplementary table 2 is marker genes
	### supplementary table 4 is DEGs, can't figure out how to download
	###########

[lcolladotor]

@kmaynard12 is going to work on this. @lahuuki, I wrote https://github.com/LieberInstitute/spatialDLPFC/tree/main/code/analysis/10_clinical_gene_set_enrichment in such a way that you would need to make 2 new scripts. One for extracting the gene IDs (ENSEMBL IDs) from the different tables @kmaynard12 will select, then another one for computing the odds ratio + making the heatmaps.

[lcolladotor]

@lahuuki check https://jhu-genomics.slack.com/archives/C044681R5H6/p1669916306967689

[lcolladotor]

@kmaynard12 are there other case/control snRNA-seq datasets beyond the PEC ones we should be looking at?

Related to https://jhu-genomics.slack.com/archives/C01EA7VDJNT/p1673285746357859

[lcolladotor]

@lahuuki I think that we can close this issue, right?