BSBenjamin Sanchez-Lengeling

Benjamin Sanchez-Lengeling

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Benjamin Sanchez-Lengeling has a total of 190 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    kinGEMs: A robust and scalable framework for resource-constraint models through stochastic tuning of deep learning-predicted kinetic parameters

    by Rana A. Barghout, Lya Chiñas Serrano, Benjamin Sanchez-Lengeling, Radhakrishnan Mahadevan

    The construction of accurate enzyme-constrained genome-scale models (ecGEMs) remains a critical challenge in systems biology, limited by sparse kinetic data and the need for biologically meaningful representations. This work presents an integrated framework combining CPI-Pred, a deep learning model to predict kinetic parameters ( k c a t ) from sequence and compound embeddings, with kinGEMs, a pipeline to incorporate these uncertainty-aware parameters into ecGEMs for metabolic optimization. By leveraging representations at multiple scales, the approach captures sequence, structure, and kinetic data to enhance model generalizability and accuracy. Rigorous benchmarking demonstrates the framework’s capability to predict growth rates and fluxes that are consistent with experimental observations and measured proteome fractions, reduce median flux variability by several folds, and yield better-constrained metabolic models by explicitly accounting for the uncertainty inherent in ML-predicted kinetic parameters. kinGEMs successfully generated ecGEMs for 93 models spanning a phylogenetically diverse set of organisms including Gram-negative and Gram-positive bacteria, mycobacteria, parasitic protists, fungi, and mammalian cell lines. These innovations breakdown the barrier to application of ecGEMs to non-model organisms and open new avenues for metabolic engineering and synthetic biology in more industrially relevant hosts.

  • Open Access

    Do protein representations encode variant function? Diagnosing residue-level protein attributions via embedding-space gradients

    by Rana Barghout, Shengqing Wu, William Chau, Jose Manuel Barraza Chavez, Paola Driza, Radhakrishnan Mahadevan, Benjamin Sanchez-Lengeling

    A common workflow in protein engineering involves building fitness predictors on protein language models (pLMs) to estimate a variant’s fitness and interpret which residues drive it. Attribution methods such as integrated gradients address this second goal by assigning residues contributions to the variant’s predicted fitness. However, without ground truth for residue importance, it is unclear whether these attributions reflect what the predictor learned about function. Here, we introduce IGLOO , a framework that scores residue-level attributions against synthetic fitness landscapes with known per-residue importance. Across experimental axes, high predictive accuracy did not guarantee faithful attributions. We identify model interventions that more than double attribution skill scores and raise TOPSIS by 29% , with similar patterns holding in real α -amylase and β -lactamase datasets. IGLOO provides a protocol for probing what pLMs learn and a guide for choosing defaults when using residue attributions to pick sequence edits and build more trustworthy protein models.

  • Open Access

    A semantic-based community model for high-fidelity tuning of olfactory mixture distances

    by Vahid Satarifard, Laura Sisson, Yikun Han, Pedro Ilídio, Matej Hladiš, Maxence Lalis, Xuebo Song, Tiffany Yang, Wenjie Yin, Aharon Ravia, et al.

    A central goal in sensory science is to establish quantitative mappings between physical stimuli and perceptual experience. Although such mappings are well defined in vision and audition, they remain elusive in olfaction, particularly for complex odor mixtures. Here, we show that perceptual distances between odor mixtures can be predicted with high fidelity and are unexpectedly well captured by a compact semantic space derived from single-molecule representations. In the Dialogue for Reverse Engineering Assessment and Methods Olfactory Mixtures Prediction Challenge, we assembled a unified dataset of odor-mixture pairs, benchmarked predictions on a hidden test set of 46 pairs, and integrated the top-performing models into a postchallenge ensemble. This model outperformed existing state-of-the-art approaches on the hidden test set, reducing RMSE by about 33% to 0.08 and increasing Pearson correlation by 53% to 0.57, and maintained strong performance on an independent validation set of 50 newly designed mixture pairs. An ensemble, retaining only olfactory semantic features for each model included, further improved predictions, raising the Pearson correlation by 7% to 0.61 on the test set and by 15% to 0.54 on the validation set. Given that semantic features were extracted from pure molecules, it suggests that mixture perception may not require fundamentally different representational principles from single-molecule olfaction. Together, these results establish a reproducible quantitative framework for olfactory mixture perception and advance efforts to measure, model, and engineer smell.


  • Open Access

    kinGEMs: A Robust and Scalable Framework for Resource-Constraint Models through Stochastic Tuning of Deep Learning-Predicted Kinetic Parameters

    by Rana A. Barghout, Lya Chiñas Serrano, Benjamin Sanchez-Lengeling, Radhakrishnan Mahadevan

    The construction of accurate enzyme-constrained genome-scale models (ecGEMs) remains a critical challenge in systems biology, limited by sparse kinetic data and the need for biologically meaningful representations. This work presents an integrated framework combining CPI-Pred, a deep learning model to predict kinetic parameters ( k cat ) from sequence and compound embeddings, with kinGEMs, a pipeline to incorporate these uncertainty-aware parameters into ecGEMs for metabolic optimization. By leveraging representations at multiple scales, the approach captures sequence, structure, and kinetic data to enhance model generalizability and accuracy. Rigorous benchmarking demonstrates the framework’s capability to predict growth rates and fluxes that are consistent with experimental observations and measured proteome fractions, reduce median flux variability by several folds, and yield better-constrained metabolic models by explicitly accounting for the uncertainty inherent in ML-predicted kinetic parameters. kinGEMs successfully generated ecGEMs for 93 models spanning a phylogenetically diverse set of organisms including Gram-negative and Gram-positive bacteria, mycobacteria, parasitic protists, fungi, and mammalian cell lines. These innovations breakdown the barrier to application of ecGEMs to non-model organisms and open new avenues for metabolic engineering and synthetic biology in more industrially relevant hosts. 1 Author Summary: Living cells are chemical factories, and computer models of their metabolism can predict how they grow, what they consume, and what they produce, useful for everything from engineering biofuel-producing microbes to understanding disease. The more realistic versions of these models require knowing how fast each cellular enzyme works, but this information is missing for the vast majority of enzymes, especially outside of a handful of well-studied organisms like E. coli . We built kinGEMs, an automated pipeline that fills these gaps using a machine learning model to predict enzyme speeds directly from protein sequence, then automatically fine-tunes those predictions so the resulting model’s growth matches real experimental measurements. We show that models built this way better match independent, real-world data they were never tuned to match, including detailed protein-abundance measurements and a classic metabolic behavior (bacteria switching to fermentation at high growth rates) that conventional models cannot reproduce without manual intervention. We also show the approach works reasonably well across 93 different organisms and successfully reconstructs known metabolic trade-offs in an oil-producing yeast unrelated to E. coli . This provides a scalable way to build usable, enzyme-aware metabolic models for organisms that currently lack the detailed experimental data such models normally require.

  • Open Access

    Does this smell the same? Learning representations of olfactory mixtures using inductive biases

    by Gary Tom, Cher Tian Ser, Ella M Rajaonson, Stanley Lo, Hyun Suk Park, Brian K Lee, Benjamin Sanchez-Lengeling

    Olfaction—how molecules are perceived as odors to humans—is a relatively less understood sensory system compared to vision or hearing. Recently, the principal odor map (POM) was introduced to digitize the olfactory properties of single compounds. However, smells in real life are not pure single molecules, but complex mixtures of molecules, whose representations remain relatively under-explored due to limited data in olfactory mixtures. We introduce POMMix, a mixture model extension of POM which leverages mono-molecular olfactory data to build meaningful mixture representations of smells. Our model builds upon the symmetries of the problem space in a hierarchical manner: (1) graph neural networks for building mono-molecular embeddings, (2) attention mechanisms for aggregating molecular representations into mixture representations, and (3) cosine prediction heads to encode olfactory perceptual distance in the mixture embedding space. POMMix achieves state-of-the-art performance across multiple datasets. We perform comprehensive ablation studies of the components of POMMix to understand the contribution of each component. We evaluate the generalizability of the model, explore olfactory phenomena with the representations, and analyze the interpretability of the representations. Our work advances the effort to digitize olfaction, highlighting the synergy of domain expertise and deep learning in crafting mixture representations in low-data regimes.


  • Open Access

    Ranking over regression for Bayesian optimization and molecule selection

    by Gary Tom, Stanley Lo, Samantha Corapi, Alán Aspuru-Guzik, Benjamin Sanchez-Lengeling

    Bayesian optimization (BO) has become an indispensable tool for autonomous decision-making across diverse applications from autonomous vehicle control to accelerated drug and materials discovery. With the growing interest in self-driving laboratories, BO of chemical systems is crucial for machine learning guided experimental planning. Typically, BO employs a regression surrogate model to predict the distribution of unseen parts of the search space. However, for the selection of molecules, picking the top candidates with respect to a distribution, the relative ordering of their properties may be more important than their exact values. In this paper, we introduce rank-based Bayesian optimization (RBO), which utilizes a ranking model as the surrogate. We present a comprehensive investigation of RBO’s optimization performance compared to conventional BO on various chemical datasets. Our results demonstrate similar or improved optimization performance using ranking models, particularly for datasets with rough structure–property landscapes and activity cliffs. Furthermore, we observe a high correlation between the surrogate ranking ability and BO performance, and this ability is maintained even at early iterations of BO optimization when using ranking surrogate models. We conclude that RBO is an effective alternative to regression-based BO, especially for optimizing novel chemical compounds.


  • Graph Data Modeling: Molecules, Proteins, & Chemical Processes

    by José Manuel Barraza-Chavez, Rana A. Barghout, Ricardo Almada-Monter, Benjamin Sanchez-Lengeling, Adrian Jinich, Radhakrishnan Mahadevan


  • Advancing molecular machine learning representations with stereoelectronics-infused molecular graphs

    by Daniil A. Boiko, Thiago Reschützegger, Benjamin Sanchez-Lengeling, Samuel M. Blau, Gabe Gomes


  • Open Access

    A deep learning and digital archaeology approach for mosquito repellent discovery

    by Jennifer N Wei, Carlos Ruiz, Marnix Vlot, Benjamin Sanchez-Lengeling, Brian K Lee, Luuk Berning, Martijn W Vos, Rob W M Henderson, Wesley W Qian, Jacob N Sanders, et al.

    Insect-borne diseases kill > 0.5 million people annually. Currently available repellents for personal or household protection are limited in their efficacy, applicability, and safety profile. Here, we describe a machine-learning-driven high-throughput method for the discovery of novel repellent molecules. To achieve this, we digitized a large, historic dataset containing ~19,000 mosquito repellency measurements. We then trained a graph neural network (GNN) to map molecular structure and repellency. We applied this model to select 317 candidate molecules to test in parallelizable behavioral assays, quantifying repellency in multiple insect vectors of the pathogens of disease and in follow-up trials with human volunteers. The GNN approach outperformed a chemoinformatic model and produced a hit rate that increased with training data size, suggesting that both model innovation and novel data collection were integral to predictive accuracy. We identified > 10 molecules with repellency similar to or greater than the most widely used repellents. We analyzed the neural responses from the mosquito antennal (olfactory) lobe to selected repellents and found strong responses to many of the tested compounds, including those predicted to be strong repellents. Results from the antennal lobe recordings also demonstrated a correlation between the evoked responses to strong repellents and our GNN representation. This approach enables computational screening of billions of possible molecules to identify empirically tractable numbers of candidate repellents, leading to accelerated progress towards solving a global health challenge.


  • Open Access

    Ranking over Regression for Bayesian Optimization and Molecule Selection

    by Gary Tom, Stanley Lo, Samantha Corapi, Alan Aspuru-Guzik, Benjamin Sanchez-Lengeling

    Bayesian optimization (BO) has become an indispensable tool for autonomous decision-making across diverse applications from autonomous vehicle control to accelerated drug and materials discovery. With the growing interest in self-driving laboratories, BO of chemical systems is crucial for machine learning (ML) guided experimental planning. Typically, BO employs a regression surrogate model to predict the distribution of unseen parts of the search space. However, for the selection of molecules, picking the top candidates with respect to a distribution, the relative ordering of their properties may be more important than their exact values. In this paper, we introduce Rank-based Bayesian Optimization (RBO), which utilizes a ranking model as the surrogate. We present a comprehensive investigation of RBO's optimization performance compared to conventional BO on various chemical datasets. Our results demonstrate similar or improved optimization performance using ranking models, particularly for datasets with rough structure-property landscapes and activity cliffs. Furthermore, we observe a high correlation between the surrogate ranking ability and BO performance, and this ability is maintained even at early iterations of BO optimization when using ranking surrogate models. We conclude that RBO is an effective alternative to regression-based BO, especially for optimizing novel chemical compounds.