Efficient Imputation for Patch-based Missing Single-cell Data via Cluster-regularized Optimal Transport
arXiv:2601.14653v2 Announce Type: replace Abstract: Missing data in single-cell sequencing datasets poses significant challenges for extracting meaningful biological insights. However, existing imputation approaches, which often assume uniformity and data completeness, struggle to address cases with large patches of missing data. In this paper, we present CROT (Cluster-Regularized Optimal Transport), an optimal transport-based imputation algorithm designed to handle patch-based missing data in tabular formats. Our approach effectively captures the underlying data structure in the presence of s
The increasing complexity and scale of single-cell sequencing data necessitate more robust solutions for data integrity and analysis, especially with prevalent missing data patches.
Improved imputation methods for single-cell data enhance the reliability of biological insights, accelerating drug discovery, disease understanding, and synthetic biology applications.
This new algorithm, CROT, specifically addresses 'patch-based' missing data, improving the accuracy and utility of large-scale single-cell datasets for research and development.
- · Biotechnology and pharmaceutical researchers
- · Synthetic biology companies
- · AI/ML developers in life sciences
- · Genomics sequencing companies
- · Firms reliant on less accurate imputation methods
- · Research efforts with noisy single-cell data
More accurate and faster insights from single-cell genomics, leading to advanced biological modeling.
Reduced experimental costs and accelerated development timelines for new biological products and therapies.
Potential for AI-driven design and optimization of biological systems at an unprecedented scale, impacting areas like drug development and sustainable materials.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG