Showing 1–10 of 10 results for author: Wong, L

Diversity Over Scale: Whole-Slide Image Variety Enables H&E Foundation Model Training with Fewer Patches

Authors: Christoph Bosch, John K. L. Wong, Martin Paulikat, Myroslav Zapukhlyak, Bharti Arora, Manasi Aichmüller-Ratnaparkhe, Jens Baumann, Shivani Karn, Rutuja Kamble, Swapnil Karnik, Bhushan Khedkar, Serey Vathana Chhut, Witali Aswolinskiy, Christian Aichmüller

Abstract: Rapid progress in computational pathology is increasingly driven by vision foundation models pretrained on vast histopathology datasets. While recent efforts have prioritized training on an ever-larger amount of patches, we take an alternative approach focused on data diversity. Our foundation model, Athena, was initialized from a pretrained model and trained on just 115 million tissue patches, se… ▽ More Rapid progress in computational pathology is increasingly driven by vision foundation models pretrained on vast histopathology datasets. While recent efforts have prioritized training on an ever-larger amount of patches, we take an alternative approach focused on data diversity. Our foundation model, Athena, was initialized from a pretrained model and trained on just 115 million tissue patches, several times fewer than recent histopathology foundation models. Rather than relying on patch volume or complex sampling heuristics, we maximize data diversity by randomly selecting only a moderate number of patches per whole-slide image from our diverse internal repository, which spans multiple countries, institutions, and scanner types. Evaluated on a single patch-level benchmark and four slide-level downstream tasks (two molecular and two morphological), Athena approaches the state-of-the-art and even surpasses several models trained on substantially larger datasets. This indicates that diversity across whole-slide images, rather than patch quantity alone, drives learning in histopathology foundation models. △ Less

Submitted 13 November, 2025; originally announced November 2025.

People use fast, flat goal-directed simulation to reason about novel problems

Authors: Katherine M. Collins, Cedegao E. Zhang, Lionel Wong, Mauricio Barba da Costa, Graham Todd, Adrian Weller, Samuel J. Cheyette, Thomas L. Griffiths, Joshua B. Tenenbaum

Abstract: Games have long been a microcosm for studying planning and reasoning in both natural and artificial intelligence, especially with a focus on expert-level or even super-human play. But real life also pushes human intelligence along a different frontier, requiring people to flexibly navigate decision-making problems that they have never thought about before. Here, we use novice gameplay to study how… ▽ More Games have long been a microcosm for studying planning and reasoning in both natural and artificial intelligence, especially with a focus on expert-level or even super-human play. But real life also pushes human intelligence along a different frontier, requiring people to flexibly navigate decision-making problems that they have never thought about before. Here, we use novice gameplay to study how people make decisions and form judgments in new problem settings. We show that people are systematic and adaptively rational in how they play a game for the first time, or evaluate a game (e.g., how fair or how fun it is likely to be) before they have played it even once. We explain these capacities via a computational cognitive model that we call the "Intuitive Gamer". The model is based on mechanisms of fast and flat (depth-limited) goal-directed probabilistic simulation--analogous to those used in Monte Carlo tree-search models of expert game-play, but scaled down to use very few stochastic samples, simple goal heuristics for evaluating actions, and no deep search. In a series of large-scale behavioral studies with over 1000 participants and 121 two-player strategic board games (almost all novel to our participants), our model quantitatively captures human judgments and decisions varying the amount and kind of experience people have with a game--from no experience at all ("just thinking"), to a single round of play, to indirect experience watching another person and predicting how they should play--and does so significantly better than much more compute-intensive expert-level models. More broadly, our work offers new insights into how people rapidly evaluate, act, and make suggestions when encountering novel problems, and could inform the design of more flexible and human-like AI systems that can determine not just how to solve new tasks, but whether a task is worth thinking about at all. △ Less

Submitted 13 October, 2025; originally announced October 2025.

Comments: Pre-print

De novo design of high-affinity protein binders with AlphaProteo

Authors: Vinicius Zambaldi, David La, Alexander E. Chu, Harshnira Patani, Amy E. Danson, Tristan O. C. Kwan, Thomas Frerix, Rosalia G. Schneider, David Saxton, Ashok Thillaisundaram, Zachary Wu, Isabel Moraes, Oskar Lange, Eliseo Papa, Gabriella Stanton, Victor Martin, Sukhdeep Singh, Lai H. Wong, Russ Bates, Simon A. Kohl, Josh Abramson, Andrew W. Senior, Yilmaz Alguel, Mary Y. Wu, Irene M. Aspalter , et al. (7 additional authors not shown)

Abstract: Computational design of protein-binding proteins is a fundamental capability with broad utility in biomedical research and biotechnology. Recent methods have made strides against some target proteins, but on-demand creation of high-affinity binders without multiple rounds of experimental testing remains an unsolved challenge. This technical report introduces AlphaProteo, a family of machine learni… ▽ More Computational design of protein-binding proteins is a fundamental capability with broad utility in biomedical research and biotechnology. Recent methods have made strides against some target proteins, but on-demand creation of high-affinity binders without multiple rounds of experimental testing remains an unsolved challenge. This technical report introduces AlphaProteo, a family of machine learning models for protein design, and details its performance on the de novo binder design problem. With AlphaProteo, we achieve 3- to 300-fold better binding affinities and higher experimental success rates than the best existing methods on seven target proteins. Our results suggest that AlphaProteo can generate binders "ready-to-use" for many research applications using only one round of medium-throughput screening and no further optimization. △ Less

Submitted 12 September, 2024; originally announced September 2024.

Comments: 45 pages, 17 figures

People use fast, goal-directed simulation to reason about novel games

Authors: Cedegao E. Zhang, Katherine M. Collins, Lionel Wong, Mauricio Barba, Adrian Weller, Joshua B. Tenenbaum

Abstract: People can evaluate features of problems and their potential solutions well before we can effectively solve them. When considering a game we have never played, for instance, we might infer whether it is likely to be challenging, fair, or fun simply from hearing the game rules, prior to deciding whether to invest time in learning the game or trying to play it well. Many studies of game play have fo… ▽ More People can evaluate features of problems and their potential solutions well before we can effectively solve them. When considering a game we have never played, for instance, we might infer whether it is likely to be challenging, fair, or fun simply from hearing the game rules, prior to deciding whether to invest time in learning the game or trying to play it well. Many studies of game play have focused on optimality and expertise, characterizing how people and computational models play based on moderate to extensive search and after playing a game dozens (if not thousands or millions) of times. Here, we study how people reason about a range of simple but novel Connect-N style board games. We ask people to judge how fair and how fun the games are from very little experience: just thinking about the game for a minute or so, before they have ever actually played with anyone else, and we propose a resource-limited model that captures their judgments using only a small number of partial game simulations and almost no look-ahead search. △ Less

Submitted 7 February, 2025; v1 submitted 19 July, 2024; originally announced July 2024.

Comments: Accepted at CogSci 2024 as a talk

Automated Process Incorporating Machine Learning Segmentation and Correlation of Oral Diseases with Systemic Health

Authors: Gregory Yauney, Aman Rana, Lawrence C. Wong, Perikumar Javia, Ali Muftu, Pratik Shah

Abstract: Imaging fluorescent disease biomarkers in tissues and skin is a non-invasive method to screen for health conditions. We report an automated process that combines intraoral fluorescent porphyrin biomarker imaging, clinical examinations and machine learning for correlation of systemic health conditions with periodontal disease. 1215 intraoral fluorescent images, from 284 consenting adults aged 18-90… ▽ More Imaging fluorescent disease biomarkers in tissues and skin is a non-invasive method to screen for health conditions. We report an automated process that combines intraoral fluorescent porphyrin biomarker imaging, clinical examinations and machine learning for correlation of systemic health conditions with periodontal disease. 1215 intraoral fluorescent images, from 284 consenting adults aged 18-90, were analyzed using a machine learning classifier that can segment periodontal inflammation. The classifier achieved an AUC of 0.677 with precision and recall of 0.271 and 0.429, respectively, indicating a learned association between disease signatures in collected images. Periodontal diseases were more prevalent among males (p=0.0012) and older subjects (p=0.0224) in the screened population. Physicians independently examined the collected images, assigning localized modified gingival indices (MGIs). MGIs and periodontal disease were then cross-correlated with responses to a medical history questionnaire, blood pressure and body mass index measurements, and optic nerve, tympanic membrane, neurological, and cardiac rhythm imaging examinations. Gingivitis and early periodontal disease were associated with subjects diagnosed with optic nerve abnormalities (p <0.0001) in their retinal scans. We also report significant co-occurrences of periodontal disease in subjects reporting swollen joints (p=0.0422) and a family history of eye disease (p=0.0337). These results indicate cross-correlation of poor periodontal health with systemic health outcomes and stress the importance of oral health screenings at the primary care level. Our screening process and analysis method, using images and machine learning, can be generalized for automated diagnoses and systemic health screenings for other diseases. △ Less

Submitted 24 October, 2018; originally announced October 2018.

Comments: Submitted to IEEE Journal of Biomedical and Health Informatics, 2018

Multicellular self-organization of P. aeruginosa due to interactions with secreted trails

Authors: Anatolij Gelimson, Kun Zhao, Calvin K. Lee, W. Till Kranz, Gerard C. L. Wong, Ramin Golestanian

Abstract: Guided movement in response to slowly diffusing polymeric trails provides a unique mechanism for self-organization of some microorganisms. To elucidate how this signaling route leads to microcolony formation, we experimentally probe the trajectory and orientation of Pseudomonas aeruginosa that propel themselves on a surface using type IV pili motility appendages, which preferentially attach to dep… ▽ More Guided movement in response to slowly diffusing polymeric trails provides a unique mechanism for self-organization of some microorganisms. To elucidate how this signaling route leads to microcolony formation, we experimentally probe the trajectory and orientation of Pseudomonas aeruginosa that propel themselves on a surface using type IV pili motility appendages, which preferentially attach to deposited exopolysaccharides. We construct a stochastic model by analyzing single-bacterium trajectories, and show that the resulting theoretical prediction for the many-body behavior of the bacteria is in quantitative agreement with our experimental characterization of how cells explore the surface via a power law strategy. △ Less

Submitted 19 September, 2016; v1 submitted 22 July, 2016; originally announced July 2016.

Comments: 10 pages

Journal ref: Phys. Rev. Lett. 117, 178102 (2016)

Inferring synthetic lethal interactions from mutual exclusivity of genetic events in cancer

Authors: Sriganesh Srihari, Jitin Singla, Limsoon Wong, Mark A. Ragan

Abstract: Background: Synthetic lethality (SL) refers to the genetic interaction between two or more genes where only their co-alteration (e.g. by mutations, amplifications or deletions) results in cell death. In recent years, SL has emerged as an attractive therapeutic strategy against cancer: by targeting the SL partners of altered genes in cancer cells, these cells can be selectively killed while sparing… ▽ More Background: Synthetic lethality (SL) refers to the genetic interaction between two or more genes where only their co-alteration (e.g. by mutations, amplifications or deletions) results in cell death. In recent years, SL has emerged as an attractive therapeutic strategy against cancer: by targeting the SL partners of altered genes in cancer cells, these cells can be selectively killed while sparing the normal cells. Consequently, a number of studies have attempted prediction of SL interactions in human, a majority by extrapolating SL interactions inferred through large-scale screens in model organisms. However, these predicted SL interactions either do not hold in human cells or do not include genes that are (frequently) altered in human cancers, and are therefore not attractive in the context of cancer therapy. Results: Here, we develop a computational approach to infer SL interactions directly from frequently altered genes in human cancers. It is based on the observation that pairs of genes that are altered in a (significantly) mutually exclusive manner in cancers are likely to constitute lethal combinations. Using genomic copy-number and gene-expression data from four cancers, breast, prostate, ovarian and uterine (total 3980 samples) from The Cancer Genome Atlas, we identify 718 genes that are frequently amplified or upregulated, and are likely to be synthetic lethal with six key DNA-damage response (DDR) genes in these cancers. By comparing with published data on gene essentiality (~16000 genes) from ten DDR-deficient cancer cell lines, we show that our identified genes are enriched among the top quartile of essential genes in these cell lines, implying that our inferred genes are highly likely to be (synthetic) lethal upon knockdown in these cell lines. △ Less

Submitted 3 October, 2015; originally announced October 2015.

Comments: 35 pages, 7 figures

MSC Class: 92Bxx

Journal ref: Biology Direct 2015, 10:57

Methods for protein complex prediction and their contributions towards understanding the organization, function and dynamics of complexes

Authors: Sriganesh Srihari, Chern Han Yong, Ashwini Patil, Limsoon Wong

Abstract: Complexes of physically interacting proteins constitute fundamental functional units responsible for driving biological processes within cells. A faithful reconstruction of the entire set of complexes is therefore essential to understand the functional organization of cells. In this review, we discuss the key contributions of computational methods developed till date (approximately between 2003 an… ▽ More Complexes of physically interacting proteins constitute fundamental functional units responsible for driving biological processes within cells. A faithful reconstruction of the entire set of complexes is therefore essential to understand the functional organization of cells. In this review, we discuss the key contributions of computational methods developed till date (approximately between 2003 and 2015) for identifying complexes from the network of interacting proteins (PPI network). We evaluate in depth the performance of these methods on PPI datasets from yeast, and highlight challenges faced by these methods, in particular detection of sparse and small or sub- complexes and discerning of overlapping complexes. We describe methods for integrating diverse information including expression profiles and 3D structures of proteins with PPI networks to understand the dynamics of complex formation, for instance, of time-based assembly of complex subunits and formation of fuzzy complexes from intrinsically disordered proteins. Finally, we discuss methods for identifying dysfunctional complexes in human diseases, an application that is proving invaluable to understand disease mechanisms and to discover novel therapeutic targets. We hope this review aptly commemorates a decade of research on computational prediction of complexes and constitutes a valuable reference for further advancements in this exciting area. △ Less

Submitted 20 May, 2015; originally announced May 2015.

Comments: 1 Table

MSC Class: 68

A Dynamic Network Formation Model for Understanding Bacterial Self-Organization into Micro-Colonies

Authors: Luca Canzian, Kun Zhao, Gerard C. L. Wong, Mihaela van der Schaar

Abstract: We propose a general parametrizable model to capture the dynamic interaction among bacteria in the formation of micro-colonies. micro-colonies represent the first social step towards the formation of structured multicellular communities known as bacterial biofilms, which protect the bacteria against antimicrobials. In our model, bacteria can form links in the form of intercellular adhesins (such a… ▽ More We propose a general parametrizable model to capture the dynamic interaction among bacteria in the formation of micro-colonies. micro-colonies represent the first social step towards the formation of structured multicellular communities known as bacterial biofilms, which protect the bacteria against antimicrobials. In our model, bacteria can form links in the form of intercellular adhesins (such as polysaccharides) to collaborate in the production of resources that are fundamental to protect them against antimicrobials. Since maintaining a link can be costly, we assume that each bacterium forms and maintains a link only if the benefit received from the link is larger than the cost, and we formalize the interaction among bacteria as a dynamic network formation game. We rigorously characterize some of the key properties of the network evolution depending on the parameters of the system. In particular, we derive the parameters under which it is guaranteed that all bacteria will join micro-colonies and the parameters under which it is guaranteed that some bacteria will not join micro-colonies. Importantly, our study does not only characterize the properties of networks emerging in equilibrium, but it also provides important insights on how the network dynamically evolves and on how the formation history impacts the emerging networks in equilibrium. This analysis can be used to develop methods to influence on- the-fly the evolution of the network, and such methods can be useful to treat or prevent biofilm-related diseases. △ Less

Submitted 23 October, 2014; originally announced October 2014.

Cooperativity and Frustration in Protein-Mediated Parallel Actin Bundles

Authors: Homin Shin, Kirstin R. Purdy Drew, James R. Bartles, Gerard C. L. Wong, Gregory M. Grason

Abstract: We examine the mechanism of bundling of cytoskeletal actin filaments by two representative bundling proteins, fascin and espin. Small-angle X-ray studies show that increased binding from linkers drives a systematic \textit{overtwist} of actin filaments from their native state, which occurs in a linker-dependent fashion. Fascin bundles actin into a continuous spectrum of intermediate twist states… ▽ More We examine the mechanism of bundling of cytoskeletal actin filaments by two representative bundling proteins, fascin and espin. Small-angle X-ray studies show that increased binding from linkers drives a systematic \textit{overtwist} of actin filaments from their native state, which occurs in a linker-dependent fashion. Fascin bundles actin into a continuous spectrum of intermediate twist states, while espin only allows for untwisted actin filaments and fully-overtwisted bundles. Based on a coarse-grained, statistical model of protein binding, we show that the interplay between binding geometry and the intrinsic \textit{flexibility} of linkers mediates cooperative binding in the bundle. We attribute the respective continuous/discontinous bundling mechanisms of fascin/espin to differences in the stiffness of linker bonds themselves. △ Less

Submitted 6 November, 2009; v1 submitted 5 November, 2009; originally announced November 2009.

Comments: 5 pages, 3 figures, figure file has been corrected in v2

Journal ref: Phys. Rev. Lett. 103, 238102 (2009)