Abdel Rahman Alsabbagh

Abdel Rahman Alsabbagh

Auto-generated profile, based on publicly available information

Abdel Rahman Alsabbagh

Biological and Environmental Science and Engineering Division, King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2025 in Nature Methods
Verified
Journal Article
Closed access
PDF file
SKSumeer Ahmad Khan
Sumeer Ahmad Khan
XMXabier Martínez-De-Morentin
Xabier Martínez-De-Morentin
AAAbdel Rahman Alsabbagh
Abdel Rahman Alsabbagh
AMAlberto Maillo
Alberto Maillo
VLVincenzo Lagani
Vincenzo Lagani
JTJesper Tegner
Jesper Tegner

27 October 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
AAAbdel Rahman Alsabbagh
Abdel Rahman Alsabbagh
AIAlberto Maillo Ruiz de Infante
Alberto Maillo Ruiz de Infante
DGDavid Gomez-Cabrero
David Gomez-Cabrero
Narsis KianiNarsis Kiani
Narsis Kiani
SKSumeer Ahmad Khan
Sumeer Ahmad Khan
JTJesper Tegner
Jesper Tegner
With the emergence of single-cell foundation models, an important question arises: how do these models perform when trained on datasets having an imbalance in cell type distribution due to rare cell types or biased sampling? We benchmark three foundation models, scGPT, scBERT, and Geneformer, using skewed single-cell cell-type distribution for cell-type annotation. While all models had reduced performance when challenged with rare cell types, scGPT and scBERT, performed better than Geneformer. Notably, in contrast to scGPT and scBERT, Geneformer uses ordinal positions of the tokenized genes rather than actual raw gene expression values. To mitigate the effect of a skewed distribution, we find that random oversampling, but not random undersampling, improved the performance for all three foundation models. Finally, scGPT, using FlashAttention, has the fastest computational speed, whereas scBERT is more memory-efficient. We conclude that tokenization and data representation are essential areas of research, and new strategies are needed to mitigate the effects of imbalanced learning in single-cell foundation models. Code and data for reproducibility are available at https://github.com/SabbaghCodes/ImbalancedLearningForSingleCellFoundationModels .