---
title: "Deep learning two-stage framework identifies key genes and pathways in esophageal adenocarcinoma"
id: "biorxiv-14-novel-two-stage-deep-learning-based-approach-applied-to-gene-expression-data"
canonical_url: "https://medichelpline.com/clinical-feed/biorxiv-14-novel-two-stage-deep-learning-based-approach-applied-to-gene-expression-data"
content_type: "clinical_feed_article"
specialty: "Oncology"
source_name: "bioRxiv (Biomedical Preprints)"
source_url: "https://www.biorxiv.org/content/10.64898/2026.09.12.751181v1?rss=1"
published_at: "2026-09-18T12:00:00.000Z"
evidence_level: "Verified Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Deep learning two-stage framework identifies key genes and pathways in esophageal adenocarcinoma
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/biorxiv-14-novel-two-stage-deep-learning-based-approach-applied-to-gene-expression-data
- **Specialty:** [Oncology](https://medichelpline.com/clinical-feed/oncology.md)
- **Primary Source:** bioRxiv (Biomedical Preprints)
- **Source URL:** [Original Journal Publication](https://www.biorxiv.org/content/10.64898/2026.09.12.751181v1?rss=1)
- **Published At:** 2026-09-18T12:00:00.000Z
- **Evidence Rating:** Verified Feed
## Executive GIST (TL;DR)
- Researchers developed a two-stage **deep learning** framework to analyze bulk RNA-seq from esophageal tissue, aiming to identify important genes and biologically relevant pathways in esophageal adenocarcinoma. - Two datasets were analyzed: GSE234304 and GSE273848, each containing tumor and non-tumor esophageal samples. - A fully connected feedforward neural network was trained for binary classification of tumor versus non-tumor samples; two feature-selection pipelines were then applied: Neural Network followed by Support Vector Regression (**NN+SVR**) and Integrated Gradients followed by SVR (**IG+SVR**). - Genes were ranked by importance using SVR-derived weights; the top 100 genes from each approach underwent enrichment analysis with Enrichr and Metascape. - The DL-based pipelines identified more expressed genes across established esophageal cancer cell lines than baseline statistical methods (LIMMA, SAM, and the t-test). - For GSE234304, **IG+SVR** identified 9 expressed genes compared with 3 by LIMMA. For GSE273848, IG+SVR identified 11 expressed genes vs 7 by the t-test. - Key genes highlighted included **CEBPB**, **SUMO1**, **RORA**, **STAT1**, **GATA**, **OCT1**, **RUNX1**, and **NR3C1**. - Enriched biological themes included nucleoprotein maturation, collagen fibril organization, immunoglobulin-mediated immune response, immune regulation, and insulin signaling. - Authors conclude that combining neural networks with attribution-based regression yields an effective, interpretable gene-selection framework for esophageal cancer research. - Details such as specific model hyperparameters, performance metrics, sample sizes per dataset, and enrichment statistics were not reported in the source abstract.
## Clinical Analysis & Structured Key Points
Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery | bioRxiv Skip to main content New Results Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery Fahad Jamie , Turki Turki , Fawaz Alsolami , View ORCID Profile Y-h. Taguchi doi: https://doi.org/10.64898/2026.09.12.751181 Fahad Jamie 1 King Abdulaziz University; Find this author on Google Scholar Find this author on PubMed Search for this author on this site Turki Turki 1 King Abdulaziz University; Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: tturki{at}kau.edu.sa Fawaz Alsolami 1 King Abdulaziz University; Find this author on Google Scholar Find this author on PubMed Search for this author on this site Y-h. Taguchi 2 Chuo University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Y-h. Taguchi Abstract Info/History Metrics Supplementary material Preview PDF Abstract Esophageal cancer (EC) is characterized by complex transcriptional alterations and therapeutic resistance, posing challenges for traditional computational methods. In this study, we propose a deep learning (DL)-based computational framework to identify important genes and biologically relevant pathways in bulk cell RNA-seq data (GSE234304 and GSE273848), which comprise tumor and non-tumor esophageal tissue samples. A fully connected feedforward neural network was trained for binary classification, and two feature selection strategies were implemented: Neural Network followed by Support Vector Regression (NN+SVR) and Integrated Gradients combined with SVR (IG+SVR). The genes were then ranked according to their weights in SVR deriving the importance scores, and the top 100 genes were subjected to enrichment analysis using Enrichr and Metascape. The proposed DL-based approaches identified a greater number of expressed genes across established esophageal cancer cell lines than LIMMA, SAM, and the t-test did. Specifically, in the GSE234304 dataset, IG + SVR, our best method, identified a total of 9 expressed genes while the best baseline method, LIMMA, identified a total of 3 expressed genes. In terms of GSE273848 dataset, IG + SVR was also the best identifying a total of 11 expressed genes while the best baseline method, t-test, had a total of 7 expressed genes. The key genes identified included CEBPB, SUMO1, RORA, STAT1, GATA, OCT1, RUNX1, and NR3C1, as well as pathways related to nucleoprotein maturation, collagen fibril organization, the immunoglobulin-mediated immune response, immune regulation, and insulin signaling. These results show that combining neural networks and attribution-based regression creates an effective and interpretable framework for selecting genes in esophageal cancer research. Competing Interest Statement The authors have declared no competing interest. Funder Information Declared King Abdulaziz University , KAU Endowment (WAQF) King Abdulaziz University , DSR Copyright The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. Back to top Previous Next Posted September 18, 2026. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery Fahad Jamie , Turki Turki , Fawaz Alsolami , Y-h. Taguchi bioRxiv 2026.09.12.751181; doi: https://doi.org/10.64898/2026.09.12.751181 Share This Article: Copy Citation Tools Novel two-stage deep learning-based approach applied to gene expression data pertaining to esophageal adenocarcinoma boosting biological knowledge discovery Fahad Jamie , Turki Turki , Fawaz Alsolami , Y-h. Taguchi bioRxiv 2026.09.12.751181; doi: https://doi.org/10.64898/2026.09.12.751181 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Areas All Articles Animal Behavior and Cognition (8010) Biochemistry (18733) Bioengineering (14878) Bioinformatics (44394) Biophysics (22589) Cancer Biology (19717) Cell Biology (26897) Clinical Trials (138) Developmental Biology (13964) Ecology (21005) Epidemiology (2067) Evolutionary Biology (25436) Genetics (16162) Genomics (23498) Immunology (18693) Microbiology (42451) Molecular Biology (18056) Neuroscience (93420) Paleontology (700) Pathology (2977) Pharmacology and Toxicology (5093) Physiology (8111) Plant Biology (15998) Scientific Communication and Education (2095) Synthetic Biology (4559) Systems Biology (10233) Zoology (2390)
## Related Clinical Research

- [TyG-WHtR and Risk of Incident Thyroid Nodules: Prospective Cohort Analysis](https://medichelpline.com/clinical-feed/pubmed-42897960.md) (DOI: 10.1080/07853890.2026.2748230)
- [Prostate cancer whole-genome profiling in African American men identifies novel drivers](https://medichelpline.com/clinical-feed/plos-one-12-personalized-genomic-landscape-of-prostate-cancer-in-african-american-men.md)
- [Modified Huanglian Wendan Decoction for PCOS‑IR: Network Pharmacology Links to AGE‑RAGE Signaling](https://medichelpline.com/clinical-feed/pubmed-42816162.md) (DOI: 10.1002/cbdv.71737)
- [Inflammatory Biomarkers and Risk of Postmenopausal ER-Positive Breast Cancer](https://medichelpline.com/clinical-feed/british-journal-of-cancer-0-inflammatory-biomarkers-and-risk-of-postmenopausal-oestrogen-receptor-positive.md)
- [Autologous DC-CIK Cell Therapy Combined with Multidisciplinary Care in Advanced Pancreatic Cancer:](https://medichelpline.com/clinical-feed/frontiers-in-immunology-17-multidisciplinary-treatment-incorporating-autologous-dc-cik-cell-therapy.md)

## Navigation
- [← Back to Oncology Feed](https://medichelpline.com/clinical-feed/oncology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.