De novo transformer modeling improves recovery of genetic cell types from sparse single-cell RNA sequencing
- 1Temple University
Abstract
Single-cell RNA sequencing (scRNA-seq) simultaneously provides gene-expression profiles and genetic variants from individual cells, creating an opportunity to relate cellular phenotypes to their somatic evolutionary histories. However, delineating genetic types (GTs) from scRNA-seq remains difficult because most bases remain undetected in each cellular genome and are accompanied by false-positive and false-negative base calls. We evaluated some existing phylogenetic and imputation methods for GT inference using one simulated and two empirical tumor datasets. We found that the extreme sparsity of cell-variant (CV) alignments produced by scRNA-seq data prevented reliable GT recovery, particularly when multiple GTs were present. This led us to explore a transformer-based genotype imputation system (STICI), which has been shown to reduce sparsity in germline genomes. However, STICI was trained on germline sequence variation in the human population, making it inapplicable for imputing idiosyncratic somatic variation in tumor genomes. We tested the hypothesis that a STICI model trained on observed bases in a given CV alignment can predict missing bases in that dataset. We found that sample-specific STICI (ssSTICI) models predicted many missing bases with high imputation probabilities, and the resulting CV alignments placed the constituent cells of each GT closer together in the inferred cellular phylogenies. Therefore, sample-specific deep learning models show strong potential to reduce CV alignment sparsity, paving the way for integrative studies of sequence and expression evolution using scRNA-seq and similar technologies.
Related
- π-MNovo improves de novo peptide sequencing through microbial-domain adaptation and evidence-guided candidate selection
- CRISPR/Cas9 knockout of PtaAGL16 improves post-drought recovery and shoot vitality in hybrid poplar
- Genotyping by Sequencing (GBS) uncovers genetic diversity and population structure for salicornia breeding programs in Iran
- Integrated optimization of experimental and computational workflows improves genome recovery in long-read gut metagenomics
- Genetic diversity of Legionella species in culture-negative clinical and environmental specimens by sequencing the 23S-5S ribosomal intergenic spacer region
BibTeX
@misc{craig2026novo,
title = {{De novo transformer modeling improves recovery of genetic cell types from sparse single-cell RNA sequencing}},
author = {J. M. Craig and M. E. Mowlaei and S. Miura and X. Shi and S. Kumar},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.03.742620},
url = {https://www.biorxiv.org/content/10.64898/2026.08.03.742620v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - De novo transformer modeling improves recovery of genetic cell types from sparse single-cell RNA sequencing
AU - J. M. Craig
AU - M. E. Mowlaei
AU - S. Miura
AU - X. Shi
AU - S. Kumar
PY - 2026
DA - 2026/08/03
PB - bioRxiv
DO - 10.64898/2026.08.03.742620
UR - https://www.biorxiv.org/content/10.64898/2026.08.03.742620v2
AB - Single-cell RNA sequencing (scRNA-seq) simultaneously provides gene-expression profiles and genetic variants from individual cells, creating an opportunity to relate cellular phenotypes to their somatic evolutionary histories. However, delineating genetic types (GTs) from scRNA-seq remains difficult because most bases remain undetected in each cellular genome and are accompanied by false-positive and false-negative base calls. We evaluated some existing phylogenetic and imputation methods for GT inference using one simulated and two empirical tumor datasets. We found that the extreme sparsity of cell-variant (CV) alignments produced by scRNA-seq data prevented reliable GT recovery, particularly when multiple GTs were present. This led us to explore a transformer-based genotype imputation system (STICI), which has been shown to reduce sparsity in germline genomes. However, STICI was trained on germline sequence variation in the human population, making it inapplicable for imputing idiosyncratic somatic variation in tumor genomes. We tested the hypothesis that a STICI model trained on observed bases in a given CV alignment can predict missing bases in that dataset. We found that sample-specific STICI (ssSTICI) models predicted many missing bases with high imputation probabilities, and the resulting CV alignments placed the constituent cells of each GT closer together in the inferred cellular phylogenies. Therefore, sample-specific deep learning models show strong potential to reduce CV alignment sparsity, paving the way for integrative studies of sequence and expression evolution using scRNA-seq and similar technologies.
ER -