Deconvolution reveals cell-type-specific transcriptomic changes in the aging mouse brain.

Mounting evidence highlights the crucial role of aging in the pathogenesis of Alzheimer's disease (AD). We have previously explored human apoE-targeted replacement mice across different ages and identified distinct molecular pathways driven by aging. However, the specific contribution of different brain cell types to the gene modules underlying these pathways remained elusive. To bridge this knowledge gap, we employed a computational deconvolution approach to examine cell-type-specific gene expression profiles in major brain cell types, including astrocytes (AS), microglia (MG), oligodendroglia (OG), neurons (NEU), and vascular cells (VC). Our findings revealed that immune module genes were predominantly expressed in MG, OG, and VC. The lipid metabolism module genes were primarily expressed in AS, MG, and OG. The mitochondria module genes showed prominent expression in VC, and the synapse module genes were primarily expressed in NEU and VC. Furthermore, we identified intra- and inter-cell-type interactions among these module genes and validated their aging-associated expression changes using published single cell studies. Our study dissected bulk brain transcriptomics data at the cellular level, providing a closer examination of the cell-type contributions to the molecular pathways driven by aging.

Projecting RNA measurements onto single cell atlases to extract cell type-specific expression profiles using scProjection.

Multi-modal single cell RNA assays capture RNA content as well as other data modalities, such as spatial cell position or the electrophysiological properties of cells. Compared to dedicated scRNA-seq assays however, they may unintentionally capture RNA from multiple adjacent cells, exhibit lower RNA sequencing depth compared to scRNA-seq, or lack genome-wide RNA measurements. We present scProjection, a method for mapping individual multi-modal RNA measurements to deeply sequenced scRNA-seq atlases to extract cell type-specific, single cell gene expression profiles. We demonstrate several use cases of scProjection, including identifying spatial motifs from spatial transcriptome assays, distinguishing RNA contributions from neighboring cells in both spatial and multi-modal single cell assays, and imputing expression measurements of un-measured genes from gene markers. scProjection therefore combines the advantages of both multi-modal and scRNA-seq assays to yield precise multi-modal measurements of single cells.

Protocol for vision transformer-based evaluation of drug potency using images processed by an optimized Sobel operator.

Conventional approaches for screening anticancer drugs rely on chemical reactions, which are time consuming, labor intensive, and costly. Here, we present a protocol for label-free and high-throughput assessment of drug efficacy using a vision transformer and a Conv2D. We describe the steps for cell culture, drug treatment, data collection, and preprocessing. We then detail the building of deep learning models and their use to predict drug potency. This protocol can be adapted for screening chemicals that affect the density or morphological features of cells. For complete details on the use and execution of this protocol, please refer to Wang et al.1.

SIC50: Determining drug inhibitory concentrations using a vision transformer and an optimized Sobel operator.

As a measure of cytotoxic potency, half-maximal inhibitory concentration (IC50) is the concentration at which a drug exerts half of its maximal inhibitory effect against target cells. It can be determined by various methods that require applying additional reagents or lysing the cells. Here, we describe a label-free Sobel-edge-based method, which we name SIC50, for the evaluation of IC50. SIC50 classifies preprocessed phase-contrast images with a state-of-the-art vision transformer and allows for the continuous assessment of IC50 in a faster and more cost-efficient manner. We have validated this method using four drugs and 1,536-well plates and also built a web application. We anticipate that this method will assist in the high-throughput screening of chemical libraries (e.g., small-molecule drugs, small interfering RNA [siRNA], and microRNA and drug discovery).

Investigation of Drosophila fruitless neurons that express Dpr/DIP cell adhesion molecules.

Drosophila reproductive behaviors are directed by fruitless neurons. A reanalysis of genomic studies shows that genes encoding dpr and DIP immunoglobulin superfamily (IgSF) members are expressed in fru P1 neurons. We find that each fru P1 and dpr/DIP (fru P1 ∩ dpr/DIP) overlapping expression pattern is similar in both sexes, but there are dimorphisms in neuronal morphology and cell number. Behavioral studies of fru P1 ∩ dpr/DIP perturbation genotypes indicate that the mushroom body functions together with the lateral protocerebral complex to direct courtship behavior. A single-cell RNA-seq analysis of fru P1 neurons shows that many DIPs have high expression in a small set of neurons, whereas the dprs are often expressed in a larger set of neurons at intermediate levels, with a myriad of dpr/DIP expression combinations. Functionally, we find that perturbations of sex hierarchy genes and of DIP-ε change the sex-specific morphologies of fru P1 ∩ DIP-α neurons.