Machine Learning & AI Algorithm Development for Biological Networks

The contract seeks the design and training of advanced machine learning and artificial intelligence models, specifically focusing on graph neural networks and temporal models, to reconstruct dynamic biological networks from sparse and incomplete experimental data. The goal is to develop algorithms capable of inferring complex biological interactions such as gene regulatory pathways or protein signaling dynamics, even when data is limited or noisy. This work targets high-impact challenges in systems biology and requires innovative approaches that combine domain expertise with cutting-edge computational techniques to yield interpretable and predictive network models. This is a small business set-aside under the SBIR/STTR program, limited to organizations with fewer than 500 employees, and falls under NAICS code 541721 for research and development in the physical, engineering, and life sciences. The contract is issued by the Defense Advanced Research Projects Agency under the U.S. Department of Defense and is structured as a subcontract. The solicitation was posted on June 24, 2026, with a response deadline of July 22, 2026. While specific performance location and point of contact details are not provided, the work is expected to align with broader DoD objectives in biological systems analysis, potentially supporting defense-relevant applications in health, biosecurity, or medical countermeasures.