BMSIP scRNAseq Analysis and Benchmarking

Analysis of scRNA-seq samples using Seurat and benchmarking of various industry standard data integration tools.

The docs folder has the poster which was presented summarising the work done over the period of the internship, as well as 2 presentations during BMSIP meetings.

Deliverables:

• Analysing and compiling results for 5 scRNASeq mouse samples using Seurat.
• Analysing and compiling results for 10 scRNASeq monkey samples using Seurat.
• Running and optimising novel graph-based method for scRNASeq (scGND) [ 1 ].
• Running tools in R and python for comparative study with the new tool actively being developed in the lab. Current tools which have been shortlisted include: MNN, fastMNN, scmap, limma, scMerge, LIGER, scGEN, MMD-ResNet, ZINB-WaVE.

Reference:

• Liu, Y. C., Zou, A., Lu, S. L., Lee, J. H., Wang, J., & Zhang, C. (2024). scGND: Graph neural diffusion model enhances single-cell RNA-seq analysis. bioRxiv, 2024-01.