Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets

Kewei Li; Yusi Fan; Yaqing Liu; Hongmei Liu; Gongyou Zhang; Meiyu Duan; Lan Huang; Fengfeng Zhou

doi:10.3791/66030

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and...

JoVE Journal
Biology

A subscription to JoVE is required to view this content. Sign in or start your free trial.

JoVE Journal Biology

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets

Full Text

1,360 Views

03:37 min

March 1, 2024

DOI: 10.3791/66030-v

Kewei Li¹, Yusi Fan¹, Yaqing Liu¹, Hongmei Liu², Gongyou Zhang², Meiyu Duan¹, Lan Huang¹, Fengfeng Zhou¹

¹College of Computer Science and Technology, and Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education,Jilin University, ²School of Biology and Engineering,Guizhou Medical University

Overview

This study addresses the intricate interactions among genes related to disease, focusing on the identification of dark biomarkers often overlooked by traditional methods. The proposed mqTrans view allows for a new understanding of these biomarkers, which exhibit differential expression compared to conventional transcriptomic analyses.

Key Study Components

Research Area

Genetic interactions related to disease.
Identification and analysis of biomarkers.
Improvement of diagnostic approaches.

Background

Genes show complex interdependencies in disease contexts.
Conventional methods often miss dark biomarkers.
Existing medical literature supports the prevalence of these biomarkers.

Methods Used

Creation of a virtual environment named Health Model in Python.
Utilization of a reference model trained on healthy samples.
Implementation of feature selection algorithms for mqTrans features.

Main Results

Identification of 221 dark biomarkers from 3,062 features.
Dark biomarkers showed differential mqTrans values but not differential mRNA expression.
The approach successfully highlighted shortcomings in traditional biomarker detection.

Conclusions

The study provides a novel methodology for unraveling dark biomarkers.
This approach could enhance biomarker screening efficiency and accuracy.

Frequently Asked Questions

What are dark biomarkers?

Dark biomarkers are genes that exhibit significant changes in expression in specific analyses but are not detected in traditional methods.

How does the mqTrans view differ from conventional methods?

The mqTrans view allows for the identification of biomarkers that do not show differential expression in standard transcriptomic analyses.

What technological requirements are needed for this study?

Users will need to create a Python virtual environment, and install packages like PyTorch and torch geometric components.

Why are healthy samples important in this research?

Healthy samples are crucial for establishing a reference model that helps in identifying deviations in disease samples.

What challenges were faced during the research?

The primary challenge was managing small sample sizes across different disease types.

What implications could this research have?

This research could potentially expedite the process of biomarker screening and lead to more precise diagnostics in disease treatment.

How many dark biomarkers were found in this study?

The study found a total of 221 dark biomarkers among the analyzed features.

Here, we introduce a protocol for converting transcriptomic data into a mqTrans view, enabling the identification of dark biomarkers. While not differentially expressed in conventional transcriptomic analyses, these biomarkers exhibit differential expression in the mqTrans view. The approach serves as a complementary technique to traditional methods, unveiling previously overlooked biomarkers.

Our research focuses on the intergenetic interactions of disease. We found that many genes have complex intertwined relationships with each other. By exploring these relationships, we aim in making more precise diagnosis in disease treatment and management.

We found that many dark biomarkers with traditional combinational methods ignored, are supported by many medical literatures. This indicates that approach can further assist biologists in reducing time cycle for biomarker screening. The biggest challenge during the experiment was addressing the issue of small sample sizes of different disease types.

View the full transcript and gain access to thousands of scientific videos