The presentation will address how TRIMTECH is leveraging its proprietary platform of small molecule TRIMTAC degraders to recruit the novel E3 ligase TRIM21 to the surface of intracellular protein aggregates. Selective degradation of multimeric forms of proteins is achieved via a unique clustering activation mechanism (ligase to ligase; proximity) that spares the monomeric protein, demonstrating the potential of this next generation targeted protein degradation approach for the development of targeted therapies for proteophathic aggregates in neurological disorders.

ARID1B is a core subunit of the SWI/SNF (BAF) chromatin-remodeling complex and a key regulator of gene accessibility. Its lack of ligandable pockets has long rendered ARID1B an undruggable target. Cancers with ARID1A loss-of-function mutations—including endometrial, ovarian, and gastric tumors—show a strong synthetic lethal dependency on ARID1B, highlighting its therapeutic potential. Using structure-based design and binder discovery, we identified the first selective ARID1B-binding chemical series and developed VHL- and CRBN-recruiting degraders that induce ARID1B loss. These findings establish targeted ARID1B degradation as a promising strategy for ARID1A-mutant cancers and provide a framework for drugging previously intractable proteins.

My laboratory develops scalable, quantitative platforms to study targeted protein degradation. By combining fluorescent stability reporters, pooled CRISPR screening, and high throughput flow cytometry, we uncover diverse small molecule induced degradation mechanisms and expand target scope. A major focus is CRBN, a highly plastic E3 ligase. Using large libraries and ultrasensitive degron discovery methods, we define rules of substrate recognition and identify targets often missed by conventional proteomics.

Current extracellular targeted protein degradation (eTPD) strategies primarily rely on recycling receptors and lysosomal trafficking for internalization and degradation. Here, we developed bispecific antibodies that recruit membrane-bound proteases to proteins of interest, enabling their “degradation” them via enzymatic shedding. Additionally, the induced proteolysis releases soluble ligands that may influence downstream cellular processes. This approach provides a new mechanism of eTPD and broadens the scope of antibody-based therapeutics.

ALK inhibitors have improved outcomes in ALK+ non-small cell lung cancer but remain limited by off-target toxicity, and resistance mutations. This work describes the discovery of TRI-611, a brain-penetrant CRL4CRBN molecular glue degrader that addresses resistance by recruiting ALK via a degron distal to the orthosteric ATP binding site. TRI-611 drives tumor regression in pre-clinical mouse xenograft models with daily oral dosing and is currently in Phase I/II clinical trials.

The central challenge in medicinal chemistry is making optimal decisions on which compounds to design and advance, rather than just predicting molecular properties. We describe an integrated framework for program decision making that combines ADMET machine learning, physics-based methods, PBPK models, AI retrosynthesis, and AI chemistry assistants. This system organizes the Design-Make-Test-Analyze (DMTA) cycle around the clinically meaningful endpoint of predicted human pharmacokinetics and efficacious dose, enabling consistent, high-quality decision making. Case studies illustrate its use in lead generation and lead optimization campaigns.

GlueFinder accelerates molecular glue discovery by systematically mining the Protein Data Bank for ligandable pockets adjacent to protein–protein interfaces that can nucleate glue-stabilized ternary complexes with E3 ligases. Benchmarking on solved dimers supports its ability to recover known and propose new glues. Applied to EGFR, HER2, and KRAS, GlueFinder predicts candidate glues recruiting 24, 111, and 148 distinct E3 ligases, respectively, and can induce non-native EGFR complexes.

We present a screening strategy for the discovery of both degradative and non-degradative macrocyclic molecular glues. Highlights will include development of assays for the discovery of kinase- and phosphatase-recruiting ligands that do no inhibit these enzymes (preferable for glue development) from one bead one compound libraries; description of a fragment-based drug development-like workflow for macrocyclic glue discovery that does not require the synthesis of extremely large libraries and an application of this technology to the creation of phosphatase recruiting chimeras (PHORCs) that engage the PTP1B phosphatase.

We will discuss how we apply state-of-the-art high throughput proteomics to discover molecular glue hits and apply AlphaFold 3-like folding software to predict the structures of molecule glue induced ternary complexes.

Recent advances in peptide display technologies enable rapid identification of high‑affinity macrocyclic peptides, yet their poor membrane permeability often limits therapeutic utility. I present a “Peptide‑to‑Small Molecule” strategy that translates peptide binding information into drug‑like small molecules through pharmacophore extraction and structure‑guided de novo design. Using some case studies, I demonstrate how this approach/concept enables discovery of cell‑active small‑molecule inhibitors for traditionally challenging targets.

FGFR2 alterations drive several cancers, including gastric, breast, and intrahepatic cholangiocarcinoma (ICC), but current FGFR inhibitors show limited benefit. We developed biparatopic antibodies that bind FGFR2, demonstrated strong potency in FGFR2-driven and fusion-positive ICC models, and also showed activity against resistant FGFR2 mutations, supporting their therapeutic promise for FGFR2-driven cancers.

We are developing a next-generation small molecule–drug conjugate (SMDC) designed to overcome the penetration limitations of conventional ADCs. By targeting a tumor-associated antigen enriched in the tumor microenvironment, this novel modality enables deep tumor infiltration and potent anti-tumor activity in complex TME settings.