Know what's in a genome, and where it came from.
Microbe-IQ builds AI systems that identify dangerous genetic elements in sequenced microbial genomes and determine whether they arose naturally or through engineering, giving biodefense and biosurveillance teams a fast, evidence-based way to triage risk.
The Approach
When a new infection or outbreak strain is sequenced, responders need to know quickly whether it carries genetic material worth investigating, and manual expert review is too slow to keep pace with an active outbreak.
Microbe-IQ addresses this in two linked stages. First, our models detect candidate foreign genetic elements in a genome, such as plasmids, insertion sequences, and prophages, and characterize their likely cargo and origin. Second, we classify whether that acquisition was natural or engineered, using molecular evidence at the insertion site to distinguish deliberate genetic engineering from natural gene transfer.
The underlying platform builds on lineage-aware genomic machine learning developed at Georgia Tech, validated on thousands of Pseudomonas aeruginosa genomes to predict source environment directly from genome-wide sequence signatures.
Focus Areas
Threat identification
Detect and characterize candidate mobile genetic elements before committing resources to full forensic review.
Threat attribution
Classify whether a genetic acquisition was natural or engineered, with interpretable, evidence-based calls.
Biosecurity & biodefense
Built for federal biosurveillance, biodefense, and microbial-forensics programs responding to outbreaks.
Contact
Microbe-IQ is currently a pre-incorporation venture led by Dr. Elijah Mehlferber, developed in partnership with the Brown Lab at Georgia Tech.