A multi-gene predictive model for the radiation sensitivity of nasopharyngeal carcinoma based on machine learning
Description
In this project, a Nasopharyngeal Cancer Radiotherapy Sensitivity Scoring System (NPC-RSS) was constructed based on machine learning methods for predicting the sensitivity of nasopharyngeal cancer patients to radiation therapy. The model integrates transcriptomic data and screens key genes through a combination of multiple machine learning algorithms to build a scoring model with good predictive performance.
Files
Steps to reproduce
In this study, gene expression data of nasopharyngeal carcinoma patients were obtained and integrated through multiple data sources, including an in-hospital cohort of 34 cases (March 2017 to July 2021) collected by Pearl River Hospital. A complete bioinformatics process was used for data processing, including quality control using Trimmomatic and FastQC, genomic comparison using HISAT2, and read counting and FPKM value calculation using featureCounts. The experimental validation phase was performed using CNE2 nasopharyngeal carcinoma cell line for cell culture and radiation therapy experiments, and gene expression validation by RT-qPCR. The immune cell infiltration was further analyzed using the CIBERSORT algorithm, and cellular heterogeneity was studied in depth by combining single-cell RNA sequencing technology. Statistical analyses were performed using R software (version 4.1.2), including differential expression analysis and statistical tests for normal and non-normal distribution data to ensure the reliability and reproducibility of the study results.
Institutions
- Southern Medical University