logo
ResearchBunny Logo
Computational scoring and experimental evaluation of enzymes generated by neural networks

Biology

Computational scoring and experimental evaluation of enzymes generated by neural networks

S. R. Johnson, X. Fu, et al.

This captivating research by Sean R. Johnson and team dives deep into evaluating 20 metrics for enzyme sequence quality, revealing a significant breakthrough with the COMPSS computational filter that boosts experimental success rates by 50-150%. Discover how this work sets a benchmark for generative models and advances the field of protein engineering.

00:00
00:00
Playback language: English
Abstract
This paper evaluates 20 computational metrics for assessing the quality of enzyme sequences generated by three models: ancestral sequence reconstruction (ASR), a generative adversarial network (GAN), and a protein language model. Over 500 natural and generated sequences were expressed and purified to benchmark these metrics against in vitro enzyme activity. A computational filter, COMPSS, was developed, improving experimental success rates by 50-150%. The findings provide a benchmark for generative models and aid in selecting active variants for experimental testing, advancing protein engineering research.
Publisher
Nature Biotechnology
Published On
Apr 23, 2024
Authors
Sean R. Johnson, Xiaozhi Fu, Sandra Viknander, Clara Goldin, Sarah Monaco, Aleksej Zelezniak, Kevin K. Yang
Tags
enzyme sequences
computational metrics
generative models
protein engineering
experimental success rates

Related Publications

Explore these studies to deepen your understanding of the subject.

Distinct neural mechanisms of social orienting and mentalizing revealed by independent measures of neural and eye movement typicality
Psychology

Distinct neural mechanisms of social orienting and mentalizing revealed by independent measures of neural and eye movement typicality

M. Ramot, C. Walsh, et al.

Advancing diagnostic performance and clinical usability of neural networks via adversarial training and dual batch normalization
Medicine and Health

Advancing diagnostic performance and clinical usability of neural networks via adversarial training and dual batch normalization

T. Han, S. Nebelung, et al.

Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials
Engineering and Technology

Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials

M. Dai, M. F. Demirel, et al.

Conversion of levoglucosan into glucose by the coordination of four enzymes through oxidation, elimination, hydration, and reduction
Biology

Conversion of levoglucosan into glucose by the coordination of four enzymes through oxidation, elimination, hydration, and reduction

Y. Kuritani, K. Sato, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny