Journal: BMC Genomics

Article Title: Integrating gene expression and imaging data across Visium capture areas with visiumStitched

doi: 10.1186/s12864-024-10991-y

Figure Lengend Snippet: Experimental design to generate spatially-resolved transcriptomics (SRT) data from postmortem human brain across three Visium capture arrays. A Schematics illustrating a coronal human brain hemisphere at the level of the anterior striatum. Inset showing an illustration of the lateral side of the brain with a vertical red line depicting the location of the coronal slab. CC - cerebral cortex, C - caudate nucleus, P - putamen, WM - white matter. B Raw coronal brain slab, dissected tissue block with an indicated score line, and H&E staining of the 10 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upmu$$\end{document} μ m section taken from the same block. C A diagram depicting how tissue sections were arranged on the Visium slide with capture arrays A1 (blue) and D1 (yellow) arranged partially overlapping, while capture arrays D1 (yellow) and C1 (red) arranged adjacent to each other across a score line

Article Snippet: The high-resolution images obtained on the 10x Genomics Visium platform were processed using VistoSeg [ ], a MATLAB -based pipeline that integrates gene expression data with histological data.

Techniques: Blocking Assay, Staining

Journal: BMC Genomics

Article Title: Integrating gene expression and imaging data across Visium capture areas with visiumStitched

doi: 10.1186/s12864-024-10991-y

Figure Lengend Snippet: Defining array coordinates for stitched data. A The transformed pixel coordinates for donor Br2719 are plotted, colored by capture area. At the overlap of capture areas V13B23-283_A1 and V13B23-283_D1 , spot centroids do not cleanly overlap; the index of a given spot is poorly defined. B The transformed pixel coordinates for Br2719 are plotted, rounded to the nearest position on a hypothetical hexagonal, Visium-like array. At the overlap of capture areas V13B23-283_A1 and V13B23-283_D1 , spot centroids occupy the same exact positions. Array coordinates may be defined by indexing the row and column on the hypothetical Visium-like array encompassing the entire Br2719 sample. C Spots from two or more overlapping capture areas, filled in blue and outlined in green, respectively, can be assigned the same array_col and array_row values with visiumStitched , enabling using data from overlapping spots in downstream analyses that rely on the Visium hexagonal grid properties to find spot neighbors. On the left, exact spot positions are shown; on the right, positions rounded to the nearest array position are shown. The spot in orange has twelve unique neighbors at six unique array coordinates

Article Snippet: The high-resolution images obtained on the 10x Genomics Visium platform were processed using VistoSeg [ ], a MATLAB -based pipeline that integrates gene expression data with histological data.

Techniques: Transformation Assay

Journal: Human Brain Mapping

Article Title: Optimizing differential identifiability improves connectome predictive modeling of cognitive deficits from functional connectivity in Alzheimer's disease

doi: 10.1002/hbm.25448

Figure Lengend Snippet: Differential identifiability framework ( I f ). (a) For each subject, two functional connectomes (FC) matrices (restA and restB) were estimated for each half of the fMRI time‐series. (b) FC matrices were vectorized (upper triangular) and placed into a group FC matrix. (c) Principal component analysis (PCA) decomposition was performed on the group FC matrix. Each PC can be arranged as a matrix in the FC domain. (d) Individual FCs were reconstructed using different number of PCs. (e) I diff was estimated for different number of PCs (in order of explained variance) and the number of PCs maximizing I diff found

Article Snippet: We used T1‐weighted MPRAGE scans and EPI fMRI scans from the initial visit in ADNI2/GO (Philips Platforms, TR/TE = 3000/30 ms, 140 volumes, 3.3 mm thickness, see www.adni-info.org for detailed protocols) for estimation of whole‐brain FCs. fMRI scans were processed with an in‐house MATLAB and FSL based pipeline (Amico et al., ).

Techniques: Functional Assay