A Geneformer-Based Model for Identifying Early Gastric Cancer Therapeutic Targets

癌症 计算机科学 医学 内科学
作者
Xuan Liu,Guohua Li,Yanshen Liao,Zishao Zhong,Hang Wei
出处
期刊: 卷期号:: 139-143 被引量:1
标识
DOI:10.1109/asens64990.2025.11011005
摘要

Gastric cancer is a major global health issue characterized by high incidence and mortality rates. Early gastric cancer has a more favorable prognosis, yet most patients are diagnosed at an advanced stage due to subtle symptoms and limited diagnostic methods. Currently, the screening for therapeutic targets in early gastric cancer primarily relies on CRISPR-Cas9, next-generation sequencing, and advanced bioinformatics analysis, facing challenges such as off-target effects, high costs, complex data interpretation, and the elucidation of causal relationships. In this study, the Geneformer model was pre-trained on a large-scale corpus and fine-tuned on an early gastric cancer dataset to identify early gastric cancer cells. Potential therapeutic targets were identified through simulated gene perturbations and disease treatments, and their effectiveness was validated using KEGG pathway enrichment analysis and biological experimental results. After rigorous quality control and preprocessing of the data, the Geneformer model achieved excellent performance in the cell state annotation task, with an accuracy of 93.47% and an F1 score of 0.9187. Additionally, the model identified 445 potential therapeutic targets for early gastric cancer, such as TFF1, TFF2, and GKN2, which were enriched in gastric cancer-related pathways, with some targets validated by biological experiments. In summary, the proposed methodologies are capable of effectively identifying early gastric cancer cells and discovering potential therapeutic targets.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
1秒前
鑫niao发布了新的文献求助10
1秒前
Forrest完成签到,获得积分10
1秒前
Hideare发布了新的文献求助10
1秒前
Anqiang发布了新的文献求助10
1秒前
2秒前
搜集达人的应助被trx采纳,获得10
2秒前
111完成签到,获得积分20
2秒前
瘦瘦以亦发布了新的文献求助10
2秒前
852的应助被Verritis采纳,获得10
3秒前
无花果的应助被Jidi采纳,获得10
3秒前
tourist585发布了新的文献求助10
3秒前
hongyi发布了新的文献求助100
3秒前
MozzieMiao发布了新的文献求助80
4秒前
4秒前
打打的应助被留胡子的小海豚采纳,获得10
4秒前
4秒前
5秒前
地塞米松L完成签到,获得积分20
5秒前
小蘑菇的应助被DDY采纳,获得10
5秒前
5秒前
5秒前
6秒前
6秒前
6秒前
yga18完成签到,获得积分10
6秒前
molihuakai的应助被干净的乐菱采纳,获得10
6秒前
coolru的应助被糊玉米采纳,获得10
6秒前
6秒前
6秒前
7秒前
7秒前
ww完成签到 ,获得积分10
8秒前
8秒前
ding的应助被tourist585采纳,获得10
8秒前
超级无敌发布了新的文献求助10
8秒前
xxx完成签到,获得积分10
8秒前
田昆完成签到,获得积分10
9秒前
9秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Boundaries and Bodhisattvas 500
高分子基功能界面水蒸发材料与应用研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8182914
求助须知:如何正确求助?哪些是违规求助? 9572166
关于积分的说明 21823466
捐赠科研通 7779489
什么是DOI,文献DOI怎么找? 3433733
关于科研通互助平台的介绍 2557961
邀请新用户注册赠送积分活动 2479897