Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Engineering proteinase K using machine learning and synthetic genes.

Jun Liao1, Manfred K Warmuth, Sridhar Govindarajan

  • 1Department of Computer Science, University of California, Santa Cruz, CA 95064, USA. liaojun@soe.ucsc.edu

BMC Biotechnology
|March 28, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Generation of high-expressing transposon-based stable pools to produce (1 + 1) and (2 + 1) bispecific common light chain antibodies.

Biotechnology progress·2026
Same author

Multiplexed antigen panel analysis identifies B cell phenotype and receptor genetic contributions to antibody breadth.

Immunity·2026
Same author

A potent Gc neutralizing antibody reveals architecture-dependent bispecific protection against SFTSV.

Emerging microbes & infections·2026
Same author

Corrigendum to "Effect of modified montmorillonite on mechanical and sustained release properties of konjac glucomannan-based hydrogel" [Int. J. Biol. Macromol. 321, Part4 (2025) 145397].

International journal of biological macromolecules·2026
Same author

Engineering stem cell-based nanotherapeutics to overcome myocardial ischemia-reperfusion injury.

Biomaterials·2026
Same author

[Naotaifang alleviates cerebral ischemia-reperfusion injury by inhibiting neuronal ferroptosis and pyroptosis through Nrf2 regulation].

Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China journal of Chinese materia medica·2026
Same journal

Dual antimicrobial and anticancer functions of artemisinin-loaded PEG<sub>400</sub>-oa nanoparticles: synergistic activity with ciprofloxacin and mechanistic insights from molecular docking.

BMC biotechnology·2026
Same journal

Editorial Expression of Concern: Hindering the biofilm of microbial pathogens and cancer cell lines development using silver nanoparticles synthesized by epidermal mucus proteins from Clarias gariepinus.

BMC biotechnology·2026
Same journal

Computer-guided enzyme engineering of PET hydrolase mutants towards improved PET affinity.

BMC biotechnology·2026
Same journal

Synthesis and application of exopolysaccharide-selenium nanocomposite produced by Pseudoduganella armeniaca ZMN-3.

BMC biotechnology·2026
Same journal

Read-mapping signatures for molecular characterization of transgenic DNA insertions in paired-end Illumina data.

BMC biotechnology·2026
Same journal

Comparative analysis of MinION and MiSeq using 16S rRNA gene amplicon sequencing in human gut microbiome.

BMC biotechnology·2026
See all related articles

A new synthetic biology method uses machine learning to engineer proteins, achieving a 20-fold improvement in proteinase K activity with fewer than 100 variants tested. This advances protein engineering for diverse applications.

Area of Science:

  • Biotechnology
  • Synthetic Biology
  • Protein Engineering

Background:

  • Traditional protein engineering methods lack high-throughput testing for predicting protein activity in specific applications.
  • Existing methods often rely on surrogate screens that may not directly correlate with desired protein functions.

Purpose of the Study:

  • To develop a novel synthetic biology approach for protein engineering that overcomes limitations of current methods.
  • To combine high-throughput gene synthesis with machine learning algorithms for efficient protein design.

Main Methods:

  • Selected 24 amino acid substitutions in proteinase K based on homologous sequence alignments.
  • Designed and synthesized 59 proteinase K variants with combinations of substitutions.
  • Tested variant activity after heat treatment and analyzed data using machine learning algorithms for iterative design.

Related Experiment Videos

Main Results:

  • Achieved a 20-fold improvement in proteinase K activity after two cycles of machine learning-guided design and testing.
  • Successfully engineered improved protein variants by testing a total of 95 variant enzymes.
  • Demonstrated the efficacy of combining gene synthesis, machine learning, and functional assays.

Conclusions:

  • Machine learning significantly reduces the number of protein variants requiring testing, enabling the use of more relevant and complex assays.
  • This approach facilitates the direct measurement of properties crucial for specific applications, such as tumor shrinkage or industrial catalysis.
  • The developed protein design algorithms represent a significant advancement towards a generic and resource-optimized protein engineering process.