4  Future perspectives: single-cell integration.

4.0.1 Alternative metrics

• Compute correlations with strongly sparse vectors?

4.0.2 Pair Single Cell with (or without) Drug fingerprints

• Single cell and compound response largest database

• Biologist perspective

  • Single-cell lineage capture across genomic modalities with CellTag-multi reveals fate-specific gene regulatory changes -> use of single-cell lineage-tracing (scLT).
  • High-resolution, noninvasive single-cell lineage tracing in mice and humans based on DNA methylation epi-mutations.

• Ming Tommy Tang lists in Single-cell LinkedIn post trendy papers and tools for multi-sample, single-cell RNAseq differential expression analysis.

4.0.2.1 Correct for Batch Effects

• DESeq2-MultiBatch: Batch Correction for Multi-Factorial RNA-seq Experiments, avalaible as open-source GH repository, from Roy, Monthony, and Torkamaneh (2025).

4.0.2.2 Drug-response

• TRADE: Transcriptome-wide analysis of differential expression in perturbation atlases, from Nadig et al. (2025). The paper notably reports how to model a compound-dose response curve with a Hill equation. Avalaible as an open-source GH repository TRADEtools. R TRADE Tutorial

4.0.3 Barcode Differential Analysis

4.0.3.1 bartools and BARtab

• Analysis of synthetic cellular barcodes in the genome and transcriptome with BARtab and bartools.

4.0.3.2 DEBRA

• Pros DEBRA
  • Better characterisation of the mean-variance deviation -> between trended or shrinkage, trended is favoured.
• Cons DEBRA:
  • DEBRA does not account for outliers expression, nor zero-inflated counts -> recommendation of glmQLFit and glmQLFTest for routine GLM-based DE analyses, from EdgeR: Explaning dispersion types to newbies.
  • Complex protocol for discarding lowly differentially expressed barcodes.
  • No available BioConductor/CRAN Repository, while latest DEBRA GitHub update dates back more than 4 years.

4.0.3.3 Combine Fold change and \(p\)-value

• Combine statistical discernibility (aka significance) with practical significance

4.0.4 Drug clustering and mapping

• Use of graph clustering approaches? Like Louvain? + multiple case studies, how to combine them (2 vials of cell lines)?
• Compare with ATC prediction and clustering: PDATC-NCPMKL-updated GH Repo.

Nadig, Ajay, Joseph M. Replogle, Angela N. Pogson, Mukundh Murthy, Steven A. McCarroll, Jonathan S. Weissman, Elise B. Robinson, and Luke J. O’Connor. 2025. ‘Transcriptome-Wide Analysis of Differential Expression in Perturbation Atlases’. Nature Genetics, April, 1–10. https://doi.org/10.1038/s41588-025-02169-3.

Roy, Julien, Adrian S. Monthony, and Davoud Torkamaneh. 2025. ‘DESeq2-MultiBatch: Batch Correction for Multi-Factorial RNA-seq Experiments’. 22 April 2025. https://doi.org/10.1101/2025.04.20.649392.