Yunji Davenport
Director Foghorn Therapeutics
Seminars
- First high-resolution structure of a human transcription factor–BAF complex revealed, mapping the PU.1–BAF binding interface
- Biochemical studies define the molecular determinants of PU.1–BAF interaction, a key driver in lineage-specific transcription
- Identification of potential small-molecule inhibitors offers a new therapeutic strategy to disrupt aberrant TF–BAF activity in cancer
Exploring the unique hurdles of designing drugs for RNA, PROTACs, and beyond. Panellists will discuss structural and physicochemical challenges, computational and biophysical tool limitations, permeability and delivery barriers, and the need for novel screening and optimization strategies to unlock these emerging therapeutic modalities.
Discussion Points:
- What structural and physicochemical features make RNA, macrocycles, and PROTACs especially challenging compared to small molecules?
- How can computational chemistry, structural biology, and biophysics be adapted to better interrogate these modalities?
- What strategies are most promising for improving permeability, stability, and delivery of large or complex molecules?
- How can we build robust optimization workflows to balance potency, selectivity, and drug-like properties across different modalities?
Despite rapid advances in structural techniques, computational models, and biophysical tools, many discovery programs still operate in disciplinary silos. This panel will explore how to weave together structural biology, biophysics, computational chemistry, and CADD to generate a holistic, data-driven view of targets and ligands, ultimately enabling more potent, selective, and developable candidates.
- How can structural insights, computational modeling, and experimental validation be better integrated into a single design loop?
- What are the practical barriers, technical, organizational, cultural, that prevent true cross-functional integration, and how can we overcome them?
- Where has integration already delivered measurable improvements in candidate quality (potency, selectivity, developability)?
- How can AI and machine learning act as a “bridge” between disciplines, and where are the limits of that approach?
- What does an ideal, multidisciplinary discovery team look like in practice?
