Identifying Genes Linked to Endometriosis and Lupus Using Machine Learning
Author Information
Author(s): Yang Pusheng, Zhu Yiping, Miao Yaxin, Wang Tao, Liu Wenwen, Zhang Jiaxin, Ge Beilei, Sun Jing
Primary Institution: Shanghai Key Laboratory of Maternal Fetal Medicine, Shanghai Institute of Maternal-Fetal Medicine and Gynecologic Oncology, Shanghai First Maternity and Infant Hospital, School of Medicine, Tongji University
Hypothesis
Endometriosis and systemic lupus erythematosus share common molecular characteristics and diagnostic genes.
Conclusion
The study identified four co-diagnostic genes that may aid in the early diagnosis and targeted therapy for endometriosis and systemic lupus erythematosus.
Supporting Evidence
- Four co-diagnostic genes (PMP22, QSOX1, REV3L, SP110) were identified for endometriosis and SLE.
- The AUC value of the four co-diagnostic genes was higher than 0.85 in both cohorts.
- Functional enrichment analysis showed significant pathways related to DNA replication and cell adhesion.
- Tributyrin was found to act on all four co-diagnostic genes.
Takeaway
Researchers found four genes that can help doctors diagnose two diseases, endometriosis and lupus, more easily.
Methodology
The study used GEO datasets to identify differentially expressed genes and constructed prediction signatures using LASSO logistic regression.
Potential Biases
Potential biases due to reliance on existing datasets and limited clinical validation.
Limitations
The study relied on public databases with a small sample size for clinical validation and requires further in vivo and in vitro experiments.
Participant Demographics
The study included samples from women with endometriosis and systemic lupus erythematosus.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website