Proceedings of the
The Nineteenth International Conference on Computational Intelligence and Security (CIS 2023)
December 1 – 4, 2023, Haikou, China
The Algorithm of Gene Expression Based on Machine Learning
1College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou, China.
2School of Computer and Information Engineering, Shanghai Polytechnic University, Shanghai, China.
ABSTRACT
The prediction of cell ratio in tissue is a hot issue in the study of cell heterogeneity. Single-cell RNA sequence can reveal the expression of genes in the genome of the single-cell level. The paper presents a Lfhlzd algorithm, use convolutional neural networks in cellular deconvolution. We uses PBMC data for experiments. The experiment results show that it can avoids the method of solving the overdetermined equation to predict the cell type ratio. We compare the performance of Lfhlzd with existing other deconvolution methods. The algorithm can quantify the content of the tissue cell population, the hidden features of the network extracted in the RNA sequencing data are more robust to data noise, without the need for complex data preprocessing.
Keywords: RNA sequences, Algorithms, Gene expression, Deconvolution, Machine learning.
Download PDF
1College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou, China.
2School of Computer and Information Engineering, Shanghai Polytechnic University, Shanghai, China.
ABSTRACT
The prediction of cell ratio in tissue is a hot issue in the study of cell heterogeneity. Single-cell RNA sequence can reveal the expression of genes in the genome of the single-cell level. The paper presents a Lfhlzd algorithm, use convolutional neural networks in cellular deconvolution. We uses PBMC data for experiments. The experiment results show that it can avoids the method of solving the overdetermined equation to predict the cell type ratio. We compare the performance of Lfhlzd with existing other deconvolution methods. The algorithm can quantify the content of the tissue cell population, the hidden features of the network extracted in the RNA sequencing data are more robust to data noise, without the need for complex data preprocessing.
Keywords: RNA sequences, Algorithms, Gene expression, Deconvolution, Machine learning.
Download PDF