Peter O. Fedichev
Research Subjects
Research Interests
Peter O. Fedichev has a total of 239 co-authors. Here are the top co-authors by number of shared publications:
- Timothy V. Pyrkov
Gero LLC
11 shared publications - Andrei E. Tarkhov
Division of Genetics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts 02115, USA
10 shared publications - Andrei V. Gudkov
Roswell Park Comprehensive Cancer Center, Buffalo, NY 14263, USA
7 shared publications G. V. Shlyapnikov
FOM Institute for Atomic and Molecular Physics, Kruislaan 407, 1098SJ Amsterdam, The Netherlands
7 shared publications- Konstantin Avchaciov
GERO PTE. LTD., 133 Cecil Street 14-01 Keck Seng Tower, Singapore 069535
7 shared publications L. I. Menshikov
Faculty of Bioengineering and Bioinformatics Gero LLC Russia. Novokuznetskaya 24/2, 4) Moscow
7 shared publications
Latest Publications
- Open Access
HARVEST: Unlocking the Dark Bioactivity Data of Pharmaceutical Patents via Agentic AI
Pharmaceutical patents contain vast Structure–Activity Relationship tables documenting protein– ligand binding data. While technically public, this information remains computationally inaccessible and effectively dark, trapped in bulky documents that no existing database has systematically captured. We present HARVEST, a multi-agent large language model pipeline that autonomously extracts structured bioactivity records from USPTO patent archives at $0.11 per document. Applied to 164,877 patents, HARVEST produced 3.15 million activity records, recovering 326,342 unique scaffolds and 967 protein targets absent from BindingDB. This pipeline completed in under a week a task that would otherwise require over 55 years of continuous expert labor. Automated extraction achieves 80% agreement with human curated corpus of US patents from BindingDB, a conservative lower bound given identified errors within the reference data. We further introduce H-Bench, a structurally guaranteed held-out benchmark built from this recovered data. Evaluation of the leading open-source model Boltz-2 on H-Bench reveals a two-dimensional generalization gap: performance degrades both on novel scaffolds and on uncharacterized protein targets, exposing fundamental limitations of models trained on existing public repositories.
Open AccessHARVEST: Unlocking the Dark Bioactivity Data of Pharmaceutical Patents via Agentic AI
Pharmaceutical patents contain vast Structure–Activity Relationship tables documenting protein– ligand binding data. While technically public, this information remains computationally inaccessible and effectively dark, trapped in bulky documents that no existing database has systematically captured. We present HARVEST, a multi-agent large language model pipeline that autonomously extracts structured bioactivity records from USPTO patent archives at $0.11 per document. Applied to 164,877 patents, HARVEST produced 3.15 million activity records, recovering 326,342 unique scaffolds and 967 protein targets absent from BindingDB. This pipeline completed in under a week a task that would otherwise require over 55 years of continuous expert labor. Automated extraction achieves 80% agreement with human curated corpus of US patents from BindingDB, a conservative lower bound given identified errors within the reference data. We further introduce H-Bench, a structurally guaranteed held-out benchmark built from this recovered data. Evaluation of the leading open-source model Boltz-2 on H-Bench reveals a two-dimensional generalization gap: performance degrades both on novel scaffolds and on uncharacterized protein targets, exposing fundamental limitations of models trained on existing public repositories.
- Open Access
Thermodynamic scaling of canine aging and reversible clock deceleration by a reverse transcriptase inhibitor
Aging was tracked in a cohort of 99 “retired” sled dogs over four years to characterize the latent dynamics of physiological decline. Animals were randomized to receive either placebo or the reverse transcriptase inhibitor lamivudine for ∼30 months. We employed a variational autoregressive model to integrate longitudinal blood parameters and DNA methylation (DNAm) profiles. The model defines Biological Age (BA) as a signature of irreversible damage with Poissonian statistics, a feature conserved across mammalian scales. Lamivudine modulated age-independent latent variables and temporarily decelerated the DNAm clock in females, but these effects were reversible upon treatment discontinuation and did not alter the long-term BA trajectory. Critically, we show that physiological fluctuations are governed by a single systemic factor – an effective or phenotypic temperature representing an emergent (macroscopic) property. We show that while the rate of damage accumulation (the BA slope) is independent of this temperature, actuarial aging parameters (initial mortality and the Gompertz exponent) are strongly temperature dependent. This supports a model where mortality arises from effective activation across a protective free energy barrier that erodes with age. These findings identify phenotypic temperature as an important control variable governing the kinetics of organism-level failure, offering a compelling target for interventions aimed at extending healthspan by “squaring” the survival curve.
Open AccessThermodynamic scaling of canine aging and reversible clock deceleration by a reverse transcriptase inhibitor
Aging was tracked in a cohort of 99 “retired” sled dogs over four years to characterize the latent dynamics of physiological decline. Animals were randomized to receive either placebo or the reverse transcriptase inhibitor lamivudine for ∼30 months. We employed a variational autoregressive model to integrate longitudinal blood parameters and DNA methylation (DNAm) profiles. The model defines Biological Age (BA) as a signature of irreversible damage with Poissonian statistics, a feature conserved across mammalian scales. Lamivudine modulated age-independent latent variables and temporarily decelerated the DNAm clock in females, but these effects were reversible upon treatment discontinuation and did not alter the long-term BA trajectory. Critically, we show that physiological fluctuations are governed by a single systemic factor – an effective or phenotypic temperature representing an emergent (macroscopic) property. We show that while the rate of damage accumulation (the BA slope) is independent of this temperature, actuarial aging parameters (initial mortality and the Gompertz exponent) are strongly temperature dependent. This supports a model where mortality arises from effective activation across a protective free energy barrier that erodes with age. These findings identify phenotypic temperature as an important control variable governing the kinetics of organism-level failure, offering a compelling target for interventions aimed at extending healthspan by “squaring” the survival curve.
- Open Access
A Minimal Model Explains Aging Regimes and Guides Intervention Strategies
Aging varies widely across species yet converges on uni-versal laws such as Gompertzian mortality. We propose a minimal framework that reduces this complexity to three macroscopic variables: the leading stress response ( z 0 ), cumulative entropic damage ( Z ), and regulatory noise strength ( D 0 ). The model predicts two regimes. In unstable species such as flies and mice, intrinsic instability drives exponential divergence of biomarkers and mortality. In stable species such as humans, linear damage accumulation steadily erodes the leading stress response, producing a hyperbolic trajectory toward a finite maximum lifespan. The framework reproduces survival curves and methylation dynamics across taxa and organizes intervention strategies into three levels: modulating stress responses, reducing noise, and slowing entropic damage, offering a roadmap for extending human healthspan and lifespan.
Open AccessA Minimal Model Explains Aging Regimes and Guides Intervention Strategies
Aging varies widely across species yet converges on uni-versal laws such as Gompertzian mortality. We propose a minimal framework that reduces this complexity to three macroscopic variables: the leading stress response ( z 0 ), cumulative entropic damage ( Z ), and regulatory noise strength ( D 0 ). The model predicts two regimes. In unstable species such as flies and mice, intrinsic instability drives exponential divergence of biomarkers and mortality. In stable species such as humans, linear damage accumulation steadily erodes the leading stress response, producing a hyperbolic trajectory toward a finite maximum lifespan. The framework reproduces survival curves and methylation dynamics across taxa and organizes intervention strategies into three levels: modulating stress responses, reducing noise, and slowing entropic damage, offering a roadmap for extending human healthspan and lifespan.
- Open Access
ProtoBind-Diff: A Structure-Free Diffusion Language Model for Protein Sequence-Conditioned Ligand Design
Designing small molecules that selectively bind to protein targets remains a central challenge in drug discovery. While recent generative models leverage 3D structural data to guide ligand generation, their applicability is limited by the sparsity and bias of structural resources. Here, we introduce ProtoBind-Diff, a structure-free masked diffusion model that conditions molecular generation directly on protein sequences via pre-trained language model embeddings. Trained on over one million active protein–ligand pairs from BindingDB, ProtoBind-Diff generates chemically valid, novel, and target-specific ligands without requiring structural supervision. In extensive benchmarking against structure-based models, ProtoBind-Diff performs competitively in docking and Boltz-1 evaluations and generalizes well to challenging targets, including those with limited training data. Despite never observing 3D information during training, its attention maps align with predicted binding residues, suggesting the model learns spatially meaningful interaction priors from sequence alone. This sequence-conditioned generation framework may unlock ligand discovery across the full proteome, including orphan, flexible, or rapidly emerging targets for which structural data are unavailable or unreliable.
Open AccessProtoBind-Diff: A Structure-Free Diffusion Language Model for Protein Sequence-Conditioned Ligand Design
Designing small molecules that selectively bind to protein targets remains a central challenge in drug discovery. While recent generative models leverage 3D structural data to guide ligand generation, their applicability is limited by the sparsity and bias of structural resources. Here, we introduce ProtoBind-Diff, a structure-free masked diffusion model that conditions molecular generation directly on protein sequences via pre-trained language model embeddings. Trained on over one million active protein–ligand pairs from BindingDB, ProtoBind-Diff generates chemically valid, novel, and target-specific ligands without requiring structural supervision. In extensive benchmarking against structure-based models, ProtoBind-Diff performs competitively in docking and Boltz-1 evaluations and generalizes well to challenging targets, including those with limited training data. Despite never observing 3D information during training, its attention maps align with predicted binding residues, suggesting the model learns spatially meaningful interaction priors from sequence alone. This sequence-conditioned generation framework may unlock ligand discovery across the full proteome, including orphan, flexible, or rapidly emerging targets for which structural data are unavailable or unreliable.
- Open Access
- 22 April 2025
- Published by Wiley in Aging Cell
- Vol. 24 (7) , e70060
AI ‐Driven Identification of Exceptionally Efficacious Polypharmacological Compounds That Extend the Lifespan of Caenorhabditis elegans
Analysis of existing lifespan‐extending geroprotective compounds suggested that polypharmacological compounds are the most effective geroprotectors, specifically those that bind multiple biogenic amine receptors. To test this hypothesis, we used graph neural networks to predict polypharmacological geroprotectors and evaluated them in Caenorhabditis elegans . Over 70% of the selected compounds extended lifespan, with effect sizes in the top 5% compared to all geroprotectors recorded in the DrugAge database. Thus, our study reveals that rationally designing polypharmacological compounds enables the design of geroprotectors with exceptional efficacy.
- 22 April 2025
- Published by Wiley in Aging Cell
- Vol. 24 (7) , e70060
Open AccessAI ‐Driven Identification of Exceptionally Efficacious Polypharmacological Compounds That Extend the Lifespan of Caenorhabditis elegans
Analysis of existing lifespan‐extending geroprotective compounds suggested that polypharmacological compounds are the most effective geroprotectors, specifically those that bind multiple biogenic amine receptors. To test this hypothesis, we used graph neural networks to predict polypharmacological geroprotectors and evaluated them in Caenorhabditis elegans . Over 70% of the selected compounds extended lifespan, with effect sizes in the top 5% compared to all geroprotectors recorded in the DrugAge database. Thus, our study reveals that rationally designing polypharmacological compounds enables the design of geroprotectors with exceptional efficacy.
- Open Access
Discovery of Thermodynamic Control Variables that Independently Regulate Healthspan and Maximum Lifespan
The question, “Can aging be modified, delayed, or reversed?” has profound social and economic implications for rapidly aging societies today. Interventions, ideally, would intercept functional decline and extend healthspan by delaying late-life morbidity (known as “squaring the curve”). These have proven elusive, but examples of differential aging in the animal world abound, suggesting aging itself is a malleable process. We present a novel multi-scale theoretical framework for entropic aging, and apply it to recently published DNA methylation data from 348 evolutionarily distant mammalian species. Our analysis identified modules or correlated DNA methylation changes associated with reversible pathway activation in key biological processes. We discovered a single species-dependent scaling factor controlling the magnitude of fluctuations across biological pathways. It acts as the organism’s “effective temperature”, quantifying intrinsic biological noise within networks and is unrelated to physical body temperature. Furthermore, we find a distinct stochastic damage signature and an associated extreme value (Gumbel) distribution of activation barriers controlling site-specific damage rates of individual CpG sites. This implies that aging is driven by rare, high-energy transitions on rugged energy landscape, most likely simultaneous and hence practically irreversible failures in highly redundant systems. While the overall rate of damage accumulation and hence the maximum lifespan does not depend on the effective temperature driving the noise in leading pathways, effective temperature does influence both initial mortality rate and the mortality rate doubling time – thereby shaping the survival curve. Lowering effective temperature must, therefore, be a promising Geroscience strategy, aimed directly at squaring the curve of aging. The example shows that targeting the thermodynamic forces driving mammalian aging may provide powerful strategies for the development of truly meaningful interventions to combat aging in humans.
Open AccessDiscovery of Thermodynamic Control Variables that Independently Regulate Healthspan and Maximum Lifespan
The question, “Can aging be modified, delayed, or reversed?” has profound social and economic implications for rapidly aging societies today. Interventions, ideally, would intercept functional decline and extend healthspan by delaying late-life morbidity (known as “squaring the curve”). These have proven elusive, but examples of differential aging in the animal world abound, suggesting aging itself is a malleable process. We present a novel multi-scale theoretical framework for entropic aging, and apply it to recently published DNA methylation data from 348 evolutionarily distant mammalian species. Our analysis identified modules or correlated DNA methylation changes associated with reversible pathway activation in key biological processes. We discovered a single species-dependent scaling factor controlling the magnitude of fluctuations across biological pathways. It acts as the organism’s “effective temperature”, quantifying intrinsic biological noise within networks and is unrelated to physical body temperature. Furthermore, we find a distinct stochastic damage signature and an associated extreme value (Gumbel) distribution of activation barriers controlling site-specific damage rates of individual CpG sites. This implies that aging is driven by rare, high-energy transitions on rugged energy landscape, most likely simultaneous and hence practically irreversible failures in highly redundant systems. While the overall rate of damage accumulation and hence the maximum lifespan does not depend on the effective temperature driving the noise in leading pathways, effective temperature does influence both initial mortality rate and the mortality rate doubling time – thereby shaping the survival curve. Lowering effective temperature must, therefore, be a promising Geroscience strategy, aimed directly at squaring the curve of aging. The example shows that targeting the thermodynamic forces driving mammalian aging may provide powerful strategies for the development of truly meaningful interventions to combat aging in humans.
- Open Access
Multiple Targets, One Goal: Compounding life‐extending effects through Polypharmacology
Analysis of lifespan‐extending compounds suggested the most effective geroprotectors target multiple biogenic amine receptors. To test this hypothesis, we used graph neural networks to predict such polypharmacological compounds and evaluated them in C. elegans . Over 70% of the selected compounds extended lifespan, with effect sizes in the top 5% compared to the DrugAge database. This reveals that rationally designing polypharmacological compounds enables the design of geroprotectors with exceptional efficacy. Key takeaways: The most effective known geroprotectors act by polypharmacological mechanisms. Graph neural networks predicted polypharmacological geroprotectors with a hit rate of 70%. The predicted polypharmacological geroprotectors are exceptionally effective. The predicted polypharmacological mechanism was experimentally confirmed. Rationally designing polypharmacological compounds results in geroprotectors with exceptional efficacy.
Open AccessMultiple Targets, One Goal: Compounding life‐extending effects through Polypharmacology
Analysis of lifespan‐extending compounds suggested the most effective geroprotectors target multiple biogenic amine receptors. To test this hypothesis, we used graph neural networks to predict such polypharmacological compounds and evaluated them in C. elegans . Over 70% of the selected compounds extended lifespan, with effect sizes in the top 5% compared to the DrugAge database. This reveals that rationally designing polypharmacological compounds enables the design of geroprotectors with exceptional efficacy. Key takeaways: The most effective known geroprotectors act by polypharmacological mechanisms. Graph neural networks predicted polypharmacological geroprotectors with a hit rate of 70%. The predicted polypharmacological geroprotectors are exceptionally effective. The predicted polypharmacological mechanism was experimentally confirmed. Rationally designing polypharmacological compounds results in geroprotectors with exceptional efficacy.
- Open Access
Differential Responses of Dynamic and Entropic Aging Factors to Longevity Interventions
Aging across most species, including mice and humans, is characterized by an exponential acceleration of mortality rates. In search for the molecular basis of this phenomenon, we analyzed DNA methylation (DNAm) changes in aging mice. Utilizing principal component analysis (PCA) on DNAm profiles, we identified a primary aging signature with an exponential age dependency, closely reflecting the Gompertz law’s description of mortality acceleration. This signature is the manifestation of the dynamic instability in the organism’s state that drives the aging process in mice. It aligns closely with regression-based aging clocks and responds to interventions such as caloric restriction and parabiosis. Additionally, we identified a linear DNAm signature, indicative of a global demethylation level. Through single-cell DNAm (scDNAm) data from aging animals, we demonstrate that this signature captures the exponential expansion of the state space volume spanned by individual cells within an aging organism, and thus quantifying linearly increasing configuration entropy, likely an irreversible process. Consistent with this interpretation, we found that neither caloric restriction (CR) nor parabiosis significantly impacts the entropic feature, reinforcing its link to irreversible damage.
Open AccessDifferential Responses of Dynamic and Entropic Aging Factors to Longevity Interventions
Aging across most species, including mice and humans, is characterized by an exponential acceleration of mortality rates. In search for the molecular basis of this phenomenon, we analyzed DNA methylation (DNAm) changes in aging mice. Utilizing principal component analysis (PCA) on DNAm profiles, we identified a primary aging signature with an exponential age dependency, closely reflecting the Gompertz law’s description of mortality acceleration. This signature is the manifestation of the dynamic instability in the organism’s state that drives the aging process in mice. It aligns closely with regression-based aging clocks and responds to interventions such as caloric restriction and parabiosis. Additionally, we identified a linear DNAm signature, indicative of a global demethylation level. Through single-cell DNAm (scDNAm) data from aging animals, we demonstrate that this signature captures the exponential expansion of the state space volume spanned by individual cells within an aging organism, and thus quantifying linearly increasing configuration entropy, likely an irreversible process. Consistent with this interpretation, we found that neither caloric restriction (CR) nor parabiosis significantly impacts the entropic feature, reinforcing its link to irreversible damage.
- Open Access
Hybrid quantum-classical machine learning for generative chemistry and drug design
Deep generative chemistry models emerge as powerful tools to expedite drug discovery. However, the immense size and complexity of the structural space of all possible drug-like molecules pose significant obstacles, which could be overcome with hybrid architectures combining quantum computers with deep classical networks. As the first step toward this goal, we built a compact discrete variational autoencoder (DVAE) with a Restricted Boltzmann Machine (RBM) of reduced size in its latent layer. The size of the proposed model was small enough to fit on a state-of-the-art D-Wave quantum annealer and allowed training on a subset of the ChEMBL dataset of biologically active compounds. Finally, we generated 2331 novel chemical structures with medicinal chemistry and synthetic accessibility properties in the ranges typical for molecules from ChEMBL. The presented results demonstrate the feasibility of using already existing or soon-to-be-available quantum computing devices as testbeds for future drug discovery applications.
Open AccessHybrid quantum-classical machine learning for generative chemistry and drug design
Deep generative chemistry models emerge as powerful tools to expedite drug discovery. However, the immense size and complexity of the structural space of all possible drug-like molecules pose significant obstacles, which could be overcome with hybrid architectures combining quantum computers with deep classical networks. As the first step toward this goal, we built a compact discrete variational autoencoder (DVAE) with a Restricted Boltzmann Machine (RBM) of reduced size in its latent layer. The size of the proposed model was small enough to fit on a state-of-the-art D-Wave quantum annealer and allowed training on a subset of the ChEMBL dataset of biologically active compounds. Finally, we generated 2331 novel chemical structures with medicinal chemistry and synthetic accessibility properties in the ranges typical for molecules from ChEMBL. The presented results demonstrate the feasibility of using already existing or soon-to-be-available quantum computing devices as testbeds for future drug discovery applications.
- Open Access
- 22 May 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-8
Hybrid quantum-classical machine learning for generative chemistry and drug design
Deep generative chemistry models emerge as powerful tools to expedite drug discovery. However, the immense size and complexity of the structural space of all possible drug-like molecules pose significant obstacles, which could be overcome with hybrid architectures combining quantum computers with deep classical networks. As the first step toward this goal, we built a compact discrete variational autoencoder (DVAE) with a Restricted Boltzmann Machine (RBM) of reduced size in its latent layer. The size of the proposed model was small enough to fit on a state-of-the-art D-Wave quantum annealer and allowed training on a subset of the ChEMBL dataset of biologically active compounds. Finally, we generated 2331 novel chemical structures with medicinal chemistry and synthetic accessibility properties in the ranges typical for molecules from ChEMBL. The presented results demonstrate the feasibility of using already existing or soon-to-be-available quantum computing devices as testbeds for future drug discovery applications.
- 22 May 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-8
Open AccessHybrid quantum-classical machine learning for generative chemistry and drug design
Deep generative chemistry models emerge as powerful tools to expedite drug discovery. However, the immense size and complexity of the structural space of all possible drug-like molecules pose significant obstacles, which could be overcome with hybrid architectures combining quantum computers with deep classical networks. As the first step toward this goal, we built a compact discrete variational autoencoder (DVAE) with a Restricted Boltzmann Machine (RBM) of reduced size in its latent layer. The size of the proposed model was small enough to fit on a state-of-the-art D-Wave quantum annealer and allowed training on a subset of the ChEMBL dataset of biologically active compounds. Finally, we generated 2331 novel chemical structures with medicinal chemistry and synthetic accessibility properties in the ranges typical for molecules from ChEMBL. The presented results demonstrate the feasibility of using already existing or soon-to-be-available quantum computing devices as testbeds for future drug discovery applications.