Dataset Information

Integrating Protein Language Models and Automatic Biofoundry for Enhanced Protein Evolution

ABSTRACT: Traditional protein engineering methods, such as directed evolution, while effective, are often slow and labor-intensive. Advances in machine learning and automated biofoundry present new opportunities for optimizing these processes. This study devises a protein language model-enabled automatic evolution platform, a closed-loop system for automated protein engineering within the Design-Build-Test-Learn cycle. The protein language model ESM-2 makes zero-shot prediction of 96 variants to initiate the cycle. The biofoundry constructs and evaluates these variants, and feeds the results back to a multi-layer perceptron to train a fitness predictor, which then makes prediction of second round of 96 variants with improved fitness. With the tRNA synthetase as a model enzyme, four-rounds of evolution carried out within 10 days lead to mutants with enzyme activity improved by up to 2.4-fold. Our system significantly enhances the speed and accuracy of protein evolution, driving faster advancements in protein engineering for industrial applications.

INSTRUMENT(S): 6520A Quadrupole Time-of-Flight LC/MS

ORGANISM(S): Escherichia Coli

SUBMITTER: Haoran Yu

LAB HEAD: Haoran Yu

PROVIDER: PXD058768 | Pride | 2024-12-25

REPOSITORIES: Pride

ACCESS DATA

Dataset's files

Source:

Items per page:

1 - 5 of 13

Similar Datasets

Project description:Gene copy-number variation, which provides the raw material for the evolution of novel genes, is surprisingly widespread in natural populations. Experimental evolution studies have demonstrated an extremely high spontaneous rate of origin of gene duplications. When organisms are suboptimally adapted to their environment, gene duplication may compensate for reduced fitness by amplifying promiscuous activity of a gene, or increasing dosage of a suboptimal gene. The overarching goal of this study is to inverstigate whether CNVs constitute a common mechanism of adaptive genetic change during compensatory evolution and to further characterize the role of natural selection in dictating their evolutionary spread at a population-genomic level. Outcrossing populations of C. elegans with low fitness were evolved for >200 generations and the frequencies of CNVs in these populations were analyzed by oligonucleotide array comparative genome hybridization, quantitative PCR, and single-worm PCR. Multiple duplications and deletions were detected in intermediate to high frequencies and several lines of evidence suggest that the changes in frequency were adaptive. 1) Many copy-number changes reached high frequency, were near fixation, or were fixed in a short time. 2) Many independent duplications and deletions in high frequency harbor overlapping regions which likely include genes that are under selection for either higher or lower rates of expression. 3) The size spectrum of deuplications and deletions in the adaptive recovery populations is significantly larger than that of spontaneous copy-number variants in mutation accumulation experiments. This is expected if larger CNVs are more likely to encompass genes that are being selected for altered gene dosage. Out results validate the great potential borne by gene copy-number changes for compensatory evolution and adaptation. Experimental genome evolution of copy-number variants in 25 experimental lines compared to 5 ancestral control lines.

			Action	DRS
	10-ncAAs.zip	Other
	NCAA1.xml	Xml
	checksum.txt	Txt
	ncAA10.xml	Xml
	ncAA2.xml	Xml

Dataset Information

Integrating Protein Language Models and Automatic Biofoundry for Enhanced Protein Evolution

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets