Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.

<h4>Background</h4>Neuropathic pain (NP) can be induced by a variety of clinical conditions, such as spinal cord injury, lumbar disc herniation (LDH), lumbar spinal stenosis, diabetes, herpes zoster, and spinal cord tumors, and inflammatory stimuli. The pathogenesis of NP is extremely co...

Full description

Saved in:
Bibliographic Details
Main Authors: Yufeng He, Ye Wei, Yongxin Wang, Chunyan Ling, Xiang Qi, Siyu Geng, Yingtong Meng, Hao Deng, Qisong Zhang, Xiaoling Qin, Guanghui Chen
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0314773
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850035132325429248
author Yufeng He
Ye Wei
Yongxin Wang
Chunyan Ling
Xiang Qi
Siyu Geng
Yingtong Meng
Hao Deng
Qisong Zhang
Xiaoling Qin
Guanghui Chen
author_facet Yufeng He
Ye Wei
Yongxin Wang
Chunyan Ling
Xiang Qi
Siyu Geng
Yingtong Meng
Hao Deng
Qisong Zhang
Xiaoling Qin
Guanghui Chen
author_sort Yufeng He
collection DOAJ
description <h4>Background</h4>Neuropathic pain (NP) can be induced by a variety of clinical conditions, such as spinal cord injury, lumbar disc herniation (LDH), lumbar spinal stenosis, diabetes, herpes zoster, and spinal cord tumors, and inflammatory stimuli. The pathogenesis of NP is extremely complex. Specifically, in LDH, the herniated nucleus pulposus exerts mechanical pressure on nerve roots, triggering local inflammation and consequent NP. Anoikis, a special form of programmed cell death, is closely related to the progression of NP. In this study, we sought to clarify the molecular characteristics of anoikis-related genes in NP, providing novel insights for the diagnosis and treatment of NP.<h4>Methods</h4>We screened NP-related genes based on the GSE124272 dataset and obtained 439 anoikis-related genes from the GeneCards database. Through Least Absolute Shrinkage and Selection Operator (LASSO) and Support Vector Machine (SVM) machine learning algorithms, six key hub genes were identified: hepatocyte growth factor (HGF), matrix metalloproteinase 13 (MMP13), c-abl oncogene 1, non-receptor tyrosine kinase (ABL1), elastase neutrophil expressed (ELANE), fatty acid synthase (FASN), and long non-coding RNA (Linc00324). Functional enrichment analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), alongside Gene Set Enrichment Analysis (GSEA) and immune infiltration analysis, were performed on these hub genes. Additionally, transcription factors and potential therapeutic drugs were predicted. We also used rats to construct an NP model and validated the analyzed hub genes using hematoxylin and eosin (H&E) staining, real-time polymerase chain reaction (PCR), and Western blotting assays.<h4>Results</h4>Our data indicated that anoikis-related genes have diagnostic value in NP patients, as confirmed by experimental results. Moreover, this study elucidated the role of these genes in immune infiltration during the pathogenesis of NP and identified potential therapeutic drugs targeting these key genes.<h4>Conclusion</h4>This study further explores the pathogenesis of NP and provides certain reference value for developing targeted therapeutic strategies, thereby improving NP management.
format Article
id doaj-art-1ca39803c6fa427e96fbc2c8be8cabfb
institution DOAJ
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-1ca39803c6fa427e96fbc2c8be8cabfb2025-08-20T02:57:34ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01202e031477310.1371/journal.pone.0314773Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.Yufeng HeYe WeiYongxin WangChunyan LingXiang QiSiyu GengYingtong MengHao DengQisong ZhangXiaoling QinGuanghui Chen<h4>Background</h4>Neuropathic pain (NP) can be induced by a variety of clinical conditions, such as spinal cord injury, lumbar disc herniation (LDH), lumbar spinal stenosis, diabetes, herpes zoster, and spinal cord tumors, and inflammatory stimuli. The pathogenesis of NP is extremely complex. Specifically, in LDH, the herniated nucleus pulposus exerts mechanical pressure on nerve roots, triggering local inflammation and consequent NP. Anoikis, a special form of programmed cell death, is closely related to the progression of NP. In this study, we sought to clarify the molecular characteristics of anoikis-related genes in NP, providing novel insights for the diagnosis and treatment of NP.<h4>Methods</h4>We screened NP-related genes based on the GSE124272 dataset and obtained 439 anoikis-related genes from the GeneCards database. Through Least Absolute Shrinkage and Selection Operator (LASSO) and Support Vector Machine (SVM) machine learning algorithms, six key hub genes were identified: hepatocyte growth factor (HGF), matrix metalloproteinase 13 (MMP13), c-abl oncogene 1, non-receptor tyrosine kinase (ABL1), elastase neutrophil expressed (ELANE), fatty acid synthase (FASN), and long non-coding RNA (Linc00324). Functional enrichment analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), alongside Gene Set Enrichment Analysis (GSEA) and immune infiltration analysis, were performed on these hub genes. Additionally, transcription factors and potential therapeutic drugs were predicted. We also used rats to construct an NP model and validated the analyzed hub genes using hematoxylin and eosin (H&E) staining, real-time polymerase chain reaction (PCR), and Western blotting assays.<h4>Results</h4>Our data indicated that anoikis-related genes have diagnostic value in NP patients, as confirmed by experimental results. Moreover, this study elucidated the role of these genes in immune infiltration during the pathogenesis of NP and identified potential therapeutic drugs targeting these key genes.<h4>Conclusion</h4>This study further explores the pathogenesis of NP and provides certain reference value for developing targeted therapeutic strategies, thereby improving NP management.https://doi.org/10.1371/journal.pone.0314773
spellingShingle Yufeng He
Ye Wei
Yongxin Wang
Chunyan Ling
Xiang Qi
Siyu Geng
Yingtong Meng
Hao Deng
Qisong Zhang
Xiaoling Qin
Guanghui Chen
Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
PLoS ONE
title Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
title_full Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
title_fullStr Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
title_full_unstemmed Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
title_short Prediction and validation of anoikis-related genes in neuropathic pain using machine learning.
title_sort prediction and validation of anoikis related genes in neuropathic pain using machine learning
url https://doi.org/10.1371/journal.pone.0314773
work_keys_str_mv AT yufenghe predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT yewei predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT yongxinwang predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT chunyanling predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT xiangqi predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT siyugeng predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT yingtongmeng predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT haodeng predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT qisongzhang predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT xiaolingqin predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning
AT guanghuichen predictionandvalidationofanoikisrelatedgenesinneuropathicpainusingmachinelearning