Miles Wang-Henderson

Miles Wang-Henderson

Auto-generated profile, based on publicly available information

Miles Wang-Henderson

Terray Therapeutics

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 August 2022 in bioRxiv
Verified
Preprint
Open access
PDF file
CNChristian Niederauer
Christian Niederauer
CNChikim Nguyen
Chikim Nguyen
MWMiles Wang-Henderson
Miles Wang-Henderson
JSJohannes Stein
Johannes Stein
SSSebastian Strauss
Sebastian Strauss
KGKristina A. Ganzinger
Kristina A. Ganzinger
DNA-PAINT based single-particle tracking (DNA-PAINT-SPT) has recently significantly enhanced observation times in in vitro SPT experiments by overcoming the constraints of fluorophore photobleaching. However, with the reported implementation, only a single target can be imaged and the technique cannot be applied straight to live cell imaging. Here we report on leveraging this technique from a proof-of-principle implementation to a useful tool for the SPT community by introducing simultaneous live cell dual-colour DNA-PAINT-SPT for quantifying protein dimerisation and tracking proteins in living cell membranes, demonstrating its improved performance over single-dye SPT.

22 August 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
BKBenjamin Kaufman
Benjamin Kaufman
EWEdward C. Williams
Edward C. Williams
RPRyan Pederson
Ryan Pederson
CUCarl Underkoffler
Carl Underkoffler
ZPZahid Panjwani
Zahid Panjwani
JPJohn Parkhill
John Parkhill
A bstract: Designing a small molecule therapeutic is a challenging multi-parameter optimization problem. Key properties, such as potency, selectivity, bioavailability, and safety must be jointly optimized to deliver an effective clinical candidate. We present COATI-LDM, a novel application of latent diffusion models to the conditional generation of property-optimized, drug-like small molecules. Diffusive generation of latent molecular encodings, rather than direct diffusive generation of molecular structures, offers an appealing way to handle the small and mismatched datasets that are common for molecular properties. We benchmark various diffusion guidance schemes and sampling methods against a pre-trained autoregressive transformer and genetic algorithms to evaluate control over potency, expert preference, and various physicochemical properties. We show that conditional diffusion allows control over the properties of generated molecules, with practical and performance advantages over competing methods. We also apply the recently introduced idea of particle guidance to enhance sample diversity. We prospectively survey a panel of medicinal chemists and determine that we can conditionally generate molecules that align with their preferences via a learned preference score. Finally, we present a partial diffusion method for the local optimization of molecular properties starting from a seed molecule. Conditional generation of small molecules using latent diffusion models on molecular encodings provides a highly practical and flexible alternative to prior molecular generation schemes.