Effect of propranolol and phentolamine on myocardial necrosis after subarachnoid haemorrhage.

TLDR

High‑dimensional, small‑sample scRNA‑seq data hampers reliable gene selection, as conventional methods become unstable and impair cell clustering and annotation. The study proposes LSH‑GAN, an enhanced GAN that generates realistic samples to augment scRNA‑seq data for improved gene selection. LSH‑GAN updates the generator training with locality‑sensitive hashing to accelerate sample generation while preserving feasibility for high‑dimensional gene selection. Experiments demonstrate that LSH‑GAN markedly enhances benchmark gene‑selection performance on synthetic and four HDSS scRNA‑seq datasets, and simulations confirm its broad applicability. Software is available at https://github.com/Snehalikalall/LSH-GAN.

Abstract

<h3>Abstract</h3> High dimensional, small sample size (HDSS) scRNA-seq data presents a challenge to the gene selection task in single cell. Conventional gene selection techniques are unstable and less reliable due to the fewer number of available samples which affects cell clustering and annotation. Here, we present an improved version of generative adversarial network (GAN) called LSH-GAN to address this issue by producing new realistic samples and combining this with the original scRNA-seq data. We update the training procedure of the generator of GAN using locality sensitive hashing which speeds up the sample generation, thus maintains the feasibility of applying gene selection procedures in high dimension scRNA-seq data. Experimental results show a significant improvement in the performance of benchmark feature (gene) selection techniques on generated samples of one synthetic and four HDSS scRNA-seq data. Comprehensive simulation study ensures the applicability of the model in the feature (gene) selection domain of HDSS scRNA-seq data. <h3>Availability</h3> The corresponding software is available at https://github.com/Snehalikalall/LSH-GAN