The final, formatted version of the article will be published soon.
ORIGINAL RESEARCH article
Front. Immunol.
Sec. Inflammation
Volume 15 - 2024 |
doi: 10.3389/fimmu.2024.1493895
Predicting patients with septic shock and sepsis through analyzing whole-blood expression of NK cell-related hub genes using an advanced machine learning framework
Provisionally accepted- 1 Weihai Municipal Hospital, Weihai, Shandong Province, China
- 2 Stritch School of Medicine, Loyola University Chicago, Maywood, Illinois, United States
- 3 University of Illinois Chicago, Chicago, United States
- 4 Virginia Maryland College of Veterinary Medicine, Virginia Tech, Blacksburg, Virginia, United States
- 5 Virginia Tech, Blacksburg, Virginia, United States
- 6 Linyi People's Hospital, Linyi, Shandong Province, China
Sepsis is a critical medical condition that results in millions of deaths every year. Identifying promising biomarkers that can aid in predicting the progression of sepsis to septic shock with rapid turnaround methods remains a challenge. Recently, transcriptomics data has emerged as an advanced resource for phenotyping and endotyping various diseases, highlighting its value in predicting disease stages. Here, we retrieved four transcriptomics datasets from the GEO database, which were previously generated using peripheral blood cell samples from patients with sepsis, septic shock, and healthy controls. Through bioinformatic analysis of these transcriptomics datasets, we identified a set of six hub genes (GZMB, PRF1, KLRD1, SH2D1A, LCK, and CD247) referred to as 6-HubGss. Using this panel, we developed a novel machine learning framework, namely, SepxFindeR. We found that the 6-HubGss panel-guided SepxFindeR model exhibited perfect performance in predicting sepsis and septic shock, as well as distinguishing septic shock from sepsis. With this machine learning platform, we effectively differentiated septic shock from sepsis in a cross-database setting. Remarkably, the SepxFindeR model is compatible with 6-HubGss panel-based RT-qPCR dataset, leading to identifying newly recruited patients with sepsis and septic shock. In conclusion, our bioinformatic analysis approach has led to the discovery of a potential biomarker panel and SepxFindeR machine learning model for predicting septic shock and distinguishing it from sepsis through comprehensive transcriptome analysis techniques and expedited processing methods.
Keywords: Novel machine learning framework for diagnosis of septic shock and sepsis Sepsis, septic shock, biomarkers, prediction, Hub genes, SepxFindeR model
Received: 09 Sep 2024; Accepted: 29 Oct 2024.
Copyright: © 2024 Du, Tan, Bu, Subramanian, Geng, Wang, Xie, Wu, Zhou, Liu, Xu, Liu and Tan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Xiao-Di Tan, University of Illinois Chicago, Chicago, United States
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.