miR-AclEAI-optimized DNA logic networks and molecular beacons for precision tumor diagnostics
ATCG at Tianjin University addresses variable exosome counts in liquid biopsy. A single fluorescence signal can confuse more sample with more target. Two channels measure exosome capture and miRNA abundance; their ratio provides internal calibration.
- Team name
- ATCG 碱基互补配队
- Institution
- School of Synthetic Biology and Biomanufacturing, Tianjin University
Team media is for event showcases and reporting only. Portrait rights remain with the individuals. For concerns about portrait use, contact the organizers. We will investigate and take appropriate action.
Team Motto
“ATCG: pairing intelligence with precision ”
Background
Two-Channel Internal Calibration
Research Approach
The Probe Design Challenge
- Design Requirement
- Probe Specificity
- Design Constraints
- Sequence & Structure
- Screening Approach
- AI Multi-Objective Screening
Ratiometric detection must distinguish target miRNA from background; candidate probes therefore need sufficient target-sequence specificity.
Sequence, secondary structure, stability and nonspecific interactions jointly affect response; optimizing a single metric is insufficient.
Computational screening reduces manual trial work. Cross-specificity matrices then check candidate responses to non-target sequences.
Research Process
Development & Validation
- 01
Stage 1
Current Progress in the Cycle
Design screens probes using sequence and structure. Build and Test construct probes and measure fluorescence, background and specificity. Learn organizes the results for machine learning. Experimental data, including cross-specificity measurements, are available; the first model is being built. The matrix shows bright diagonal and low off-diagonal signals. - 02
Stage 2
Target vs. Non-Target Fluorescence
Molecular beacons fluoresce when target sequences open their hairpins. Current tests show fluorescence recovery in the experimental group and low non-target background, enabling comparison of candidate specificity. - 03
Stage 3
Start with the Scientific Question
The team knew biological experiments but initially lacked experience with ML data formats, features and model selection. Limited wet-lab data prompted changes to the initial computational plan. The team now defines the question and available data first, then selects methods and iterates between computation and experiments.
Project Evidence
Methods & Figures
Team films
