PlasmidHawk

PlasmidHawk predicts the depositing laboratory (lab-of-origin) of synthetic DNA sequences by identifying lab-specific sequence signatures in a pan-genome database.

Key Features:

• High Prediction Accuracy: Achieves 63% accuracy in predicting the depositing lab of unknown sequences, with the correct lab appearing in the top 10 candidates 80% of the time.
• Explainability and Precision: Identifies specific sub-sequences responsible for predictions, enabling interpretation of lab-of-origin determinations.
• Alignment-Based Methodology: Uses alignment-based analysis within a pan-genome database to identify unique sequence signatures associated with specific laboratories.

Scientific Applications:

• Tracking and Verification: Enables attribution and verification of synthetic DNA sequence origins.
• Biosafety and Biosecurity: Supports mitigation of misuse of synthetic biology technologies by providing origin attribution data.
• Compliance and Responsible Research: Facilitates enforcement of safety standards and investigation of provenance in research and manufacturing contexts.

Methodology:

Employs an alignment-based strategy using a comprehensive pan-genome database to identify sub-sequences indicative of specific laboratories' engineering styles or signatures.