# scFSNN: a feature selection method based on neural network for single-cell RNA-seq data

## TL;DR

This paper introduces scFSNN, a neural network-based method for selecting important genes in single-cell RNA sequencing data.

## Contribution

The novelty lies in using a neural network to automatically select features while controlling false discovery rates and adapting to data complexity.

## Key findings

- scFSNN outperforms existing methods in feature selection for scRNA-seq data.
- The method adapts to data characteristics like over-dispersion and zero-inflation.
- Simulation and real data studies confirm its strong predictive performance.

## Abstract

While single-cell RNA sequencing (scRNA-seq) allows researchers to analyze gene expression in individual cells, its unique characteristics like over-dispersion, zero-inflation, high gene-gene correlation, and large data volume with many features pose challenges for most existing feature selection methods. In this paper, we present a feature selection method based on neural network (scFSNN) to solve classification problem for the scRNA-seq data. scFSNN is an embedded method that can automatically select features (genes) during model training, control the false discovery rate of selected features and adaptively determine the number of features to be eliminated. Extensive simulation and real data studies demonstrate its excellent feature selection ability and predictive performance.

The online version contains supplementary material available at 10.1186/s12864-024-10160-1.