Biology subjects

Buchan, D.

Publications and source records attributed to Buchan, D..

4 recordsLinked to original sources

bioRxiv2025

Improving cell-type identification with generative adversarial networks-enhanced augmentation-free single-cell RNA-Seq contrastive learning

Cell-type identification plays a fundamental role in single-cell RNA-Seq analytics. Thanks to the recent success of the contrastive learning paradigm, the accuracy of automatic cell-type identification has also been improved. In this work, we propose a novel contrastive learning-based cell-type identification method, namely GAN-RCL, which conducts hard positive sample selection on real and conditional generative adversarial networks-created synthetic instances to improve the performance of augmentation-free contrastive learning. Experimental results confirm that GAN-RCL successfully outperforms other recently-proposed contrastive learning-based methods and shows the state-of-the-art predictive performance on multiple single-cell RNA-Seq cell-type identification tasks.

bioinformatics↗

bioRxiv2025

An extensive evaluation of single-cell RNA-Seq contrastivelearning generative networks for intrinsic cell-typesdistribution estimation

Contrastive learning has already been widely used to handle single-cell RNA-Seq data due to its outstanding performance in transforming original data distributions into hypersphere feature spaces. In this work, we conduct a large-scale empirical evaluation to investigate the generative encoder networks that are learned by five different state-of-the-art single-cell RNA-Seq contrastive learning methods. Unlike the conventional discriminative model-based cell-type prediction studies, this work is focused on the performance of contrastive learning-based generative encoder networks in terms of their capacity to estimate the intrinsic distributions of different cell-types - a fundamental property that directly affects the performance of any downstream single-cell RNA-Seq data analytics. The experimental results confirm that supervised contrastive learning-based encoder networks lead to better performance than self-supervised contrastive learning-based encoder networks, and the recently proposed Gaussian noise augmentation-based single-cell RNA-Seq contrastive learning method shows the best performance on estimating the intrinsic distribution of different cell-types.

bioinformatics↗

bioRxiv2024

Designing minimal E. coli genomes using variational autoencoders

Designing minimal bacterial genomes remains a key challenge in synthetic biology. There is currently a lack of efficient tools for the rapid generation of streamlined bacterial genomes, limiting research in this area. Here, using a pangenome dataset for Escherichia coli, we show that variational autoencoders with modified loss functions can successfully create minimised genomes retaining the essential genes identified in the literature. We then sampled new genomes from our fitted model and performed computational validation using an E. coli whole-cell model. We found 6 out of 100 of the sampled genomes were viable in the computer model. These underwent a minimization routine starting from the MG1655 genome giving rise to six new minimal genomes with around a 40 % reduction in size. This study proposes a rapid, machine learning-based approach for bacterial sequence generation, that could accelerate the genomic design process.

synthetic biology↗

bioRxiv2022

Improving cell-type identification with Gaussian noise-augmented single-cell RNA-seq contrastive learning

Cell-type identification is an important task for single-cell RNA-seq (scRNA-seq) data analysis. In this work, we proposed a novel Gaussian noise augmented scRNA-seq contrastive learning framework (GsRCL) to learn a type of discriminative feature representations for cell-type prediction tasks. The experimental results suggest that the feature representations learned by GsRCL successfully improved the accuracy of cell-type prediction using scRNA-seq expression profiles.

bioinformatics↗