A5036742375- Computational Drug Discovery Methods
- Genetics, Aging, and Longevity in Model Organisms
- Machine Learning in Materials Science
- Bioinformatics and Genomic Networks
- Health, Environment, Cognitive Aging
- Cancer Genomics and Diagnostics
- Epigenetics and DNA Methylation
- Fluoride Effects and Removal
- Lung Cancer Treatments and Mutations
- RNA modifications and cancer
- Genetics, Bioinformatics, and Biomedical Research
- DNA Repair Mechanisms
- Cancer Mechanisms and Therapy
- Frailty in Older Adults
- Cancer-related molecular mechanisms research
- Gene Regulatory Network Analysis
- Bone and Dental Protein Studies
- MicroRNA in disease regulation
- Ubiquitin and proteasome pathways
- Dietary Effects on Health
- Nutritional Studies and Diet
- Human Health and Disease
- Aging and Gerontology Research
- Cancer therapeutics and mechanisms
- Gene expression and cancer classification
Buck Institute for Research on Aging
2018-2025
Hong Kong Science and Technology Parks Corporation
2019-2025
Insilicos (United States)
2016-2025
Insilico Medicine (United States)
2015-2025
Zhejiang Medicine (China)
2024
OpenAI (United States)
2022
Johns Hopkins University
2014-2021
Institute of Sociology
2021
Ural Federal University
2021
Urology Foundation
2011-2020
Generative models are becoming a tool of choice for exploring the molecular space. These learn on large training dataset and produce novel structures with similar properties. Generated can be utilized virtual screening or semi-supervized predictive in downstream tasks. While there plenty generative models, it is unclear how to compare rank them. In this work, we introduce benchmarking platform called Molecular Sets (MOSES) standardize comparison models. MOSES provides testing datasets, set...
Deep learning is rapidly advancing many areas of science and technology with multiple success stories in image, text, voice video recognition, robotics, autonomous driving. In this paper we demonstrate how deep neural networks (DNN) trained on large transcriptional response data sets can classify various drugs to therapeutic categories solely based their profiles. We used the perturbation samples 678 across A549, MCF-7, PC-3 cell lines from LINCS Project linked those 12 use derived MeSH. To...
Deep generative adversarial networks (GANs) are the emerging technology in drug discovery and biomarker development. In our recent work, we demonstrated a proof-of-concept of implementing deep autoencoder (AAE) to identify new molecular fingerprints with predefined anticancer properties. Another popular model is variational (VAE), which based on neural architectures. this developed an advanced AAE for feature extraction problems, its advantages compared VAE terms (a) adjustability generating...
The increased availability of data and recent advancements in artificial intelligence present the unprecedented opportunities healthcare major challenges for patients, developers, providers regulators. novel deep learning transfer techniques are turning any about person into medical transforming simple facial pictures videos powerful sources predictive analytics. Presently, patients do not have control over access privileges to their records remain unaware true value they have. In this...
In silico modeling is a crucial milestone in modern drug design and development. Although computer-aided approaches this field are well-studied, the application of deep learning methods research area at beginning. work, we present an original neural network (DNN) architecture named RANC (Reinforced Adversarial Neural Computer) for de novo novel small-molecule organic structures based on generative adversarial (GAN) paradigm reinforcement (RL). As generator uses differentiable computer (DNC),...
Recent advances in deep learning and specifically generative adversarial networks have demonstrated surprising results generating new images videos upon request even using natural language as input. In this paper we present the first application of autoencoders (AAE) for novel molecular fingerprints with a defined set parameters. We developed 7-layer AAE architecture latent middle layer serving discriminator. As an input output uses vector binary concentration molecule. also introduced...
Aging | doi:10.18632/aging.100968. Evgeny Putin, Polina Mamoshina, Alexander Aliper, Mikhail Korzinkin, Alexey Moskalev, Kolosov, Ostrovskiy, Charles Cantor, Jan Vijg, Alex Zhavoronkov
Abstract A majority of cancers fail to respond immunotherapy with antibodies targeting immune checkpoints, such as cytotoxic T-lymphocyte antigen-4 (CTLA-4) or programmed death-1 (PD-1)/PD-1 ligand (PD-L1). Cancers frequently express transforming growth factor-β (TGFβ), which drives dysfunction in the tumor microenvironment by inducing regulatory T cells (Tregs) and inhibiting CD8 + H 1 cells. To address this therapeutic challenge, we invent bifunctional antibody–ligand traps (Y-traps)...
Modern computational approaches and machine learning techniques accelerate the invention of new drugs. Generative models can discover novel molecular structures within hours, while conventional drug discovery pipelines require months work. In this article, we propose a generative architecture, entangled conditional adversarial autoencoder, that generates based on various properties, such as activity against specific protein, solubility, or ease synthesis. We apply proposed model to generate...
In this article, we propose the deep neural network Adversarial Threshold Neural Computer (ATNC). The ATNC model is intended for de novo design of novel small-molecule organic structures. based on generative adversarial architecture and reinforcement learning. uses a Differentiable as generator has new specific block, called threshold (AT). AT acts filter between agent (generator) environment (discriminator + objective reward functions). Furthermore, to generate more diverse molecules...
The application of artificial intelligence (AI) has been considered a revolutionary change in drug discovery and development. In 2020, the AlphaFold computer program predicted protein structures for whole human genome, which remarkable breakthrough both AI applications structural biology. Despite varying confidence levels, these could still significantly contribute to structure-based design novel targets, especially ones with no or limited information. this work, we successfully applied our...
For the past several decades, research in understanding molecular basis of human muscle aging has progressed significantly. However, development accessible tissue-specific biomarkers that may be measured to evaluate effectiveness therapeutic interventions is still a major challenge. Here we present method for tracking age-related changes skeletal muscle. We analyzed publicly available gene expression profiles young and old tissue from healthy donors. Differential pathway analysis were...
Large language models utilizing transformer neural networks and other deep learning architectures demonstrated unprecedented results in many tasks previously accessible only to human intelligence. In this article, we collaborate with ChatGPT, an AI model developed by OpenAI speculate on the applications of Rapamycin, context Pascal's Wager philosophical argument commonly utilized justify belief god. response query "Write exhaustive research perspective why taking Rapamycin may be more...
Accurate and physiologically meaningful biomarkers for human aging are key to assessing antiaging therapies. Given ethnic differences in health, diet, lifestyle, behavior, environmental exposures, even average rate of biological aging, it stands reason that clocks trained on datasets obtained from specific populations more likely account these potential confounding factors, resulting an enhanced capacity predict chronological age quantify age. Here, we present a deep learning-based...
The human gut microbiome is a complex ecosystem that both affects and affected by its host status. Previous metagenomic analyses of microflora revealed associations between specific microbes age. Nonetheless there was no reliable way to tell host's age based on the community composition. Here we developed method predicting hosts' taxonomic profiles using cross-study dataset deep learning. Our best model has an architecture neural network achieves mean absolute error 5.91 years when tested...
DNA methylation aging clocks have become an invaluable tool in biogerontology research since their inception 2013. Today, a variety of machine learning approaches been tested for the purpose predicting human age based on molecular-level features. Among these, deep learning, or neural networks, is especially promising approach that has used to construct accurate using blood biochemistry, transcriptomics, and microbiomics data—feats unachieved by other algorithms. In this article, we explore...
Chemistry42 is a software platform for de novo small molecule design and optimization that integrates Artificial Intelligence (AI) techniques with computational medicinal chemistry methodologies. efficiently generates novel molecular structures optimized properties validated in both vitro vivo studies available through licensing or collaboration. the core component of Insilico Medicine's Pharma.ai drug discovery suite. also includes PandaOmics target multiomics data analysis, inClinico─a...