Digital Genetic Twins Aim to Predict Disease but Face Legal and Ethical Hurdles

Researchers and companies are developing AI-powered digital replicas of patients’ genomes to predict diseases and drug responses, while regulators and ethicists identify gaps in existing laws.

• Predictive AI’s “Doctor Twin AI” system predicts risks for over 22,000 conditions and responses to more than 210 drugs, the company stated.
• The “Evo 2” model, trained on 9 trillion DNA base pairs, identified disease-causing mutations in a breast cancer gene, as reported in Nature.
• A partnership between An Learn AI and Johnson & Johnson reduced control group size by up to one-third in Phase 3 Alzheimer’s trials, according to study results.
• A Harvard law school study concluded that current laws have structural problems for digital twins, including limited scope and conflicting intellectual property and accountability requirements.