DeepP450: Predicting Human P450 Activities of Small Molecules by Integrating Pretrained Protein Language Model and Molecular Representation DOI

Jiamin Chang,

Xiaoyu Fan, Boxue Tian

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 64(8), P. 3149 - 3160

Published: April 8, 2024

Cytochrome P450 enzymes (CYPs) play a crucial role in Phase I drug metabolism the human body, and CYP activity toward compounds can significantly affect druggability, making early prediction of substrate identification essential for therapeutic development. Here, we established deep learning model assessing potential substrates, DeepP450, by fine-tuning protein molecule pretrained models through feature integration with cross-attention self-attention layers. This exhibited high accuracy (0.92) on test set, area under receiver operating characteristic curve (AUROC) values ranging from 0.89 to 0.98 substrate/nonsubstrate predictions across nine major CYPs, surpassing current benchmarks prediction. Notably, DeepP450 uses only one predict substrates/nonsubstrates any CYPs exhibits certain generalizability novel different categories which could greatly facilitate stage design avoiding CYP-reactive compounds.

Language: Английский

From nature to industry: Harnessing enzymes for biocatalysis DOI
Rebecca Buller, Stefan Lutz, Romas J. Kazlauskas

et al.

Science, Journal Year: 2023, Volume and Issue: 382(6673)

Published: Nov. 23, 2023

Biocatalysis harnesses enzymes to make valuable products. This green technology is used in countless applications from bench scale industrial production and allows practitioners access complex organic molecules, often with fewer synthetic steps reduced waste. The last decade has seen an explosion the development of experimental computational tools tailor enzymatic properties, equipping enzyme engineers ability create biocatalysts that perform reactions not present nature. By using (chemo)-enzymatic synthesis routes or orchestrating intricate cascades, scientists can synthesize elaborate targets ranging DNA pharmaceuticals starch made vitro CO2-derived methanol. In addition, new chemistries have emerged through combination biocatalysis transition metal catalysis, photocatalysis, electrocatalysis. review highlights recent key developments, identifies current limitations, provides a future prospect for this rapidly developing technology.

Language: Английский

Citations

207
Enzyme function prediction using contrastive learning DOI
Tianhao Yu, Haiyang Cui, Jianan Canal Li

et al.

Science, Journal Year: 2023, Volume and Issue: 379(6639), P. 1358 - 1363

Published: March 31, 2023

Enzyme function annotation is a fundamental challenge, and numerous computational tools have been developed. However, most of these cannot accurately predict functional annotations, such as enzyme commission (EC) number, for less-studied proteins or those with previously uncharacterized functions multiple activities. We present machine learning algorithm named CLEAN (contrastive learning-enabled annotation) to assign EC numbers enzymes better accuracy, reliability, sensitivity compared the state-of-the-art tool BLASTp. The contrastive framework empowers confidently (i) annotate understudied enzymes, (ii) correct mislabeled (iii) identify promiscuous two more numbers-functions that we demonstrate by systematic in silico vitro experiments. anticipate this will be widely used predicting thereby advancing many fields, genomics, synthetic biology, biocatalysis.

Language: Английский

Citations

193
Machine Learning-Guided Protein Engineering DOI Creative Commons
Petr Kouba, Pavel Kohout, Faraneh Haddadi

et al.

ACS Catalysis, Journal Year: 2023, Volume and Issue: 13(21), P. 13863 - 13895

Published: Oct. 13, 2023

Recent progress in engineering highly promising biocatalysts has increasingly involved machine learning methods. These methods leverage existing experimental and simulation data to aid the discovery annotation of enzymes, as well suggesting beneficial mutations for improving known targets. The field protein is gathering steam, driven by recent success stories notable other areas. It already encompasses ambitious tasks such understanding predicting structure function, catalytic efficiency, enantioselectivity, dynamics, stability, solubility, aggregation, more. Nonetheless, still evolving, with many challenges overcome questions address. In this Perspective, we provide an overview ongoing trends domain, highlight case studies, examine current limitations learning-based We emphasize crucial importance thorough validation emerging models before their use rational design. present our opinions on fundamental problems outline potential directions future research.

Language: Английский

Citations

91
Opportunities and Challenges for Machine Learning-Assisted Enzyme Engineering DOI Creative Commons
Jason Yang, Francesca-Zhoufan Li, Frances H. Arnold

et al.

ACS Central Science, Journal Year: 2024, Volume and Issue: 10(2), P. 226 - 241

Published: Feb. 5, 2024

Enzymes can be engineered at the level of their amino acid sequences to optimize key properties such as expression, stability, substrate range, and catalytic efficiency-or even unlock new activities not found in nature. Because search space possible proteins is vast, enzyme engineering usually involves discovering an starting point that has some desired activity followed by directed evolution improve its "fitness" for a application. Recently, machine learning (ML) emerged powerful tool complement this empirical process. ML models contribute (1) discovery functional annotation known protein or generating novel with functions (2) navigating fitness landscapes optimization mappings between associated values. In Outlook, we explain how complements discuss future potential improved outcomes.

Language: Английский

Citations

70
A comprehensive review of sustainable bioremediation techniques: Eco friendly solutions for waste and pollution management DOI Creative Commons

Narendra Kuppan,

Midhila Padman,

M Mahadeva

et al.

Waste Management Bulletin, Journal Year: 2024, Volume and Issue: 2(3), P. 154 - 171

Published: July 22, 2024

Bioremediation, an advanced and environmentally sustainable technology, utilizes biological microorganisms to mitigate pollution. This review combines insights from two perspectives: one focusing on the mechanisms, applications, types of bioremediation, other examining transformative potential integrating Internet Things (IoT), Artificial Intelligence (AI), biosensors in pollution management. The first perspective delves into effectiveness bioremediation decomposing detoxifying hazardous substances, emphasizing its cost-effectiveness eco-friendliness compared conventional methods. In-situ ex-situ methods are analyzed, along with intrinsic engineered techniques, phytoremediation strategies for heavy metal removal. underscores growing importance addressing industrial effluents, contaminated soils, groundwater, future advancements expected enhance efficiency applicability. From second perspective, recent IoT, AI, explored their revolutionize waste IoT facilitates real-time monitoring remote management, AI enhances data analysis predictive modelling, contribute precise pollutant detection environmental monitoring. highlights synergistic integration these technologies, presenting smart systems feedback loops adaptive capabilities. Together, technologies offer scalable solutions mitigation, marking a significant stride towards

Language: Английский

Citations

42
Automated in vivo enzyme engineering accelerates biocatalyst optimization DOI Creative Commons
Enrico Orsi, Lennart Schada von Borzyskowski, Stephan Noack

et al.

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: April 24, 2024

Achieving cost-competitive bio-based processes requires development of stable and selective biocatalysts. Their realization through in vitro enzyme characterization engineering is mostly low throughput labor-intensive. Therefore, strategies for increasing while diminishing manual labor are gaining momentum, such as vivo screening evolution campaigns. Computational tools like machine learning further support efforts by widening the explorable design space. Here, we propose an integrated solution to challenges whereby ML-guided, automated workflows (including library generation, implementation hypermutation systems, adapted laboratory evolution, growth-coupled selection) could be realized accelerate pipelines towards superior

Language: Английский

Citations

26
Supervised machine learning in drug discovery and development: Algorithms, applications, challenges, and prospects DOI Creative Commons
George Obaido, Ibomoiye Domor Mienye, Oluwaseun Francis Egbelowo

et al.

Machine Learning with Applications, Journal Year: 2024, Volume and Issue: 17, P. 100576 - 100576

Published: July 24, 2024

Language: Английский

Citations

20
Learning from Protein Engineering by Deconvolution of Multi‐Mutational Variants DOI Creative Commons
Frank Hollmann,

Joaquin Sanchis,

Manfred T. Reetz

et al.

Angewandte Chemie International Edition, Journal Year: 2024, Volume and Issue: 63(36)

Published: June 17, 2024

Abstract This review analyzes a development in biochemistry, enzymology and biotechnology that originally came as surprise. Following the establishment of directed evolution stereoselective enzymes organic chemistry, concept partial or complete deconvolution selective multi‐mutational variants was introduced. Early experiments led to finding mutations can interact cooperatively antagonistically with one another, not just additively. During past decade, this phenomenon shown be general. In some studies, molecular dynamics (MD) quantum mechanics/molecular mechanics (QM/MM) computations were performed order shed light on origin non‐additivity at all stages an evolutionary upward climb. Data used construct unique multi‐dimensional rugged fitness pathway landscapes, which provide mechanistic insights different from traditional landscapes. Along related line, biochemists have long tested result introducing two point enzyme for reasons, followed by comparison respective double mutant so‐called cycles, showed only additive effects, but more recently also uncovered cooperative antagonistic non‐additive effects. We conclude suggestions future work, call unified overall picture epistasis.

Language: Английский

Citations

18
Biocatalytic reductive aminations with NAD(P)H-dependent enzymes: enzyme discovery, engineering and synthetic applications DOI
Bo Yuan, Dameng Yang, Ge Qu

et al.

Chemical Society Reviews, Journal Year: 2023, Volume and Issue: 53(1), P. 227 - 262

Published: Dec. 7, 2023

This review summarized NAD(P)H-dependent amine dehydrogenases and imine reductases which catalyzes asymmetric reductive amination to produce optically active amines.

Language: Английский

Citations

28
Machine Learning-Enabled Genome Mining and Bioactivity Prediction of Natural Products DOI
Yujie Yuan, Chengyou Shi, Huimin Zhao

et al.

ACS Synthetic Biology, Journal Year: 2023, Volume and Issue: 12(9), P. 2650 - 2662

Published: Aug. 22, 2023

Natural products (NPs) produced by microorganisms and plants are a major source of drugs, herbicides, fungicides. Thanks to recent advances in DNA sequencing, bioinformatics, genome mining tools, vast amount data on NP biosynthesis has been generated over the years, which increasingly exploited develop machine learning (ML) tools for discovery. In this review, we discuss latest developing applying ML exploring potential NPs that can be encoded genomic language predicting types bioactivities NPs. We also examine technical challenges associated with development application research.

Language: Английский

Citations

23