OmicVerse: a framework for bridging and deepening insights across bulk and single-cell sequencing

Area of Science:

• Genomics
• Computational Biology
• Bioinformatics

Background:

• Single-cell RNA sequencing (scRNA-seq) is limited by sequencing throughput, leading to
• omission
• of cells.
• Bulk RNA sequencing (bulk RNA-seq) data may contain information from these ostensibly omitted cells.

Purpose of the Study:

• To introduce the single cell trajectory blending from Bulk RNA-seq (BulkTrajBlend) algorithm.
• To develop a computational tool within the OmicVerse suite for restoring omitted cells in scRNA-seq data.
• To enhance the continuity and completeness of single-cell datasets.

Main Methods:

• Leveraging a Beta-Variational AutoEncoder for data deconvolution.
• Utilizing graph neural networks for the discovery of overlapping communities.
• Implementing the BulkTrajBlend algorithm for interpolating and restoring omitted cells.

Main Results:

• Successfully interpolated and restored the continuity of omitted cells within scRNA-seq datasets.
• Demonstrated the capability of BulkTrajBlend to integrate information from bulk RNA-seq.
• OmicVerse provides a toolkit for comprehensive bulk and single-cell RNA-seq analysis.

Conclusions:

• BulkTrajBlend effectively addresses the cell omission problem in scRNA-seq.
• The OmicVerse suite streamlines computational processes and enhances data visualization for RNA-seq analysis.
• This approach facilitates the extraction of significant biological insights from integrated omics data.