human genome version grch38  (OmicSoft Corporation)

Journal: bioRxiv

Article Title: Haplotype rather than single causal variants effects contribute to regulatory gene expression associations in human myeloid cells

doi: 10.1101/2025.01.30.635675

Figure Lengend Snippet: Results of Initial Screening and Validation Experiments. (A) Barplot of 87 genes tested in initial common variant experiments encompassing 4,382 gRNA across 50 transfections. A few SNPs were found in eQTL for multiple genes, so we ultimately investigated 4,814 SNP-gene regulatory relationships. Tested but nonsignificant candidates for a target gene are in salmon (n=4,272), significant candidates in maroon (n=432), nonsignificant negative controls in light grey (n=103), and significant negative controls in dark grey (n=7). (B) Heat map reveals the total number of significant candidates across all three cell types per eQTL peak following validation experiments for 76 genes brought forward. Each locus is separated into the number of peaks detected at that locus following all-but-one conditional analysis. Dark grey blocks indicate that peak number was not present at the locus. The darkest maroon indicates more than 10 significant candidates associated with that peak. (C) Custom LocusZoom plot highlighting the eQTL for GPX4 with log transformed significance on the y-axis and chromosomal position (GrCh38/hg38) on the x-axis. Significance reports the level of association with gene expression following all-but-one conditional analysis. Each point is a variant is shaded based on eCROPseq results with salmon points being all variants tested, but not significant. Maroon points indicate candidates that were validated in undifferentiated HL-60 only. (D) Custom LocusZoom for LGALS9, with same shading as in C with the addition of blue points that represent candidates that were significant in macrophage only and upwards black triangle that indicate a variant that was significant in both undifferentiated and macrophage. (E) Schematic showing four eQTL Categories and respective LD and significant candidate patterns. Darker blue shading indicates higher LD with the lead variant. Maroon points highlight significant candidates in one cell type, coral points highlight significant candidates in another cell type, and black squares highlight variants that were significant in multiple cell types.

Article Snippet: All fastq files were then aligned to a supplemented GrCh38 human reference genome (version refdata-gex-GRCh38-2020-A, 10x Genomics), which was supplemented with artificial chromosomes for each gRNA in a library.

Techniques: Variant Assay, Transfection, Transformation Assay, Expressing

Journal: bioRxiv

Article Title: Haplotype rather than single causal variants effects contribute to regulatory gene expression associations in human myeloid cells

doi: 10.1101/2025.01.30.635675

Figure Lengend Snippet: Overlapping Significant Variants Between Cell Types. (A) A) An example of Category IVa shown on a custom LocusZoom (56) plot for the eQTL of SNAPC4. Points are highlighted based on eCROPseq results with salmon points being all variants tested, but not significant. Maroon points indicate candidates that were significant in undifferentiated HL-60 only and blue in macrophage only. (B) An example of Category IVb shown on a custom LocusZoom of ERAP2, with same shading as in A. In addition, purple indicates significant candidates in neutrophil, downwards triangle indicates a variant that was significant in both undifferentiated and neutrophil, and dark grey a significant control. (C) Circos plot shading along the ring shows the number of tested candidates, but not significant (lighter color) and significant candidates (p-value < 0.05) (darker color) for undifferentiated HL-60 in maroon, macrophage in blue, and neutrophil in purple. The size of darker connections indicates the number of significant candidates detected following eCROPseq overlapping between each cell type: 18 between undifferentiated and macrophage, 14 between undifferentiated and neutrophil, and 16 between macrophage and neutrophil. Five SNPs overlap between all three cell types. The size of the larger translucent connections indicates the number of true causal candidates estimated statistically to overlap in each pairwise comparison: 121 between undifferentiated and macrophages, 102 between undifferentiated and neutrophils, and 133 between macrophages and neutrophils. (D) Primary eQTL detected for IFNAR2 with same shading as in A and B. In total we detected 7 significant candidates across the three cell types for peak 1. Specifically, one was significant in both undifferentiated HL-60 and neutrophil (rs17860115), four were specific to undifferentiated HL-60, and two were specific to macrophage. (E) Rs17860115 location highlighted in yellow revealing overlap with ATACseq peaks in undifferentiated HL-60 cells in the first track, ATACseq peaks in 120hr differentiated neutrophils in the second track, and reported cis-regulatory elements from ENCODE and NCBI in the third track. Rs17860115 overlaps with a promoter like signature reported in ENCODE and predicted silencer. The fourth track shows the position of IFNAR2 with respect to the variant. Rs17860115 falls into the 5’ UTR of the gene. (F) Primary eQTL peak for NFATC1, following eCROPseq with same shading as previously. Two candidates were found to be significant, one in macrophages (rs9748916), the other in neutrophils (rs4799052). (G) Variance explained by significant candidates (rs9748916 and rs4799052) with effect sizes of 0.5 and −0.4, together these variants explain as much variance as the lead variant (rs9962906) with effect size 0.45. All positions reported are with respect to the GrCh38/hg38 human reference genome.

Article Snippet: All fastq files were then aligned to a supplemented GrCh38 human reference genome (version refdata-gex-GRCh38-2020-A, 10x Genomics), which was supplemented with artificial chromosomes for each gRNA in a library.

Techniques: Variant Assay, Control, Comparison