Protein redesign by learning from data

Bastiaan A. Van Den Berg, Marcel J T Reinders, Jan Metske Van Der Laan, J. Roubos, Dick De Ridder

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Protein redesign methods aim to improve a desired property by carefully selecting mutations in relevant regions guided by protein structure. However, often protein structural requirements underlying biological characteristics are not well understood. Here, we introduce a methodology that learns relevant mutations from a set of proteins that have the desired property and demonstrate it by successfully improving production levels of two enzymes by Aspergillus niger, a relevant host organism for industrial enzyme production. We validated our method on two enzymes, an esterase and an inulinase, creating four redesigns with 5-45 mutations. Up to 10-fold increase in production was obtained with preserved enzyme activity for small numbers of mutations, whereas production levels and activities dropped for too aggressive redesigns. Our results demonstrate the feasibility of protein redesign by learning. Such an approach has great potential for improving production levels of many industrial enzymes and could potentially be employed for other design goals.

Original languageEnglish
Pages (from-to)281-288
Number of pages8
JournalProtein Engineering, Design and Selection
Volume27
Issue number9
DOIs
Publication statusPublished - 2014

Fingerprint

Learning
Proteins
Enzymes
Mutation
inulinase
Aspergillus niger
Aspergillus
Enzyme activity
Esterases

Keywords

  • Enzyme production levels
  • Protein redesign
  • Redesign by learning

ASJC Scopus subject areas

  • Biochemistry
  • Biotechnology
  • Bioengineering
  • Molecular Biology
  • Medicine(all)

Cite this

Van Den Berg, B. A., Reinders, M. J. T., Van Der Laan, J. M., Roubos, J., & De Ridder, D. (2014). Protein redesign by learning from data. Protein Engineering, Design and Selection, 27(9), 281-288. https://doi.org/10.1093/protein/gzu031

Protein redesign by learning from data. / Van Den Berg, Bastiaan A.; Reinders, Marcel J T; Van Der Laan, Jan Metske; Roubos, J.; De Ridder, Dick.

In: Protein Engineering, Design and Selection, Vol. 27, No. 9, 2014, p. 281-288.

Research output: Contribution to journalArticle

Van Den Berg, BA, Reinders, MJT, Van Der Laan, JM, Roubos, J & De Ridder, D 2014, 'Protein redesign by learning from data', Protein Engineering, Design and Selection, vol. 27, no. 9, pp. 281-288. https://doi.org/10.1093/protein/gzu031
Van Den Berg BA, Reinders MJT, Van Der Laan JM, Roubos J, De Ridder D. Protein redesign by learning from data. Protein Engineering, Design and Selection. 2014;27(9):281-288. https://doi.org/10.1093/protein/gzu031
Van Den Berg, Bastiaan A. ; Reinders, Marcel J T ; Van Der Laan, Jan Metske ; Roubos, J. ; De Ridder, Dick. / Protein redesign by learning from data. In: Protein Engineering, Design and Selection. 2014 ; Vol. 27, No. 9. pp. 281-288.
@article{bfa4bdcb43ed4aec9584de0c2060a077,
title = "Protein redesign by learning from data",
abstract = "Protein redesign methods aim to improve a desired property by carefully selecting mutations in relevant regions guided by protein structure. However, often protein structural requirements underlying biological characteristics are not well understood. Here, we introduce a methodology that learns relevant mutations from a set of proteins that have the desired property and demonstrate it by successfully improving production levels of two enzymes by Aspergillus niger, a relevant host organism for industrial enzyme production. We validated our method on two enzymes, an esterase and an inulinase, creating four redesigns with 5-45 mutations. Up to 10-fold increase in production was obtained with preserved enzyme activity for small numbers of mutations, whereas production levels and activities dropped for too aggressive redesigns. Our results demonstrate the feasibility of protein redesign by learning. Such an approach has great potential for improving production levels of many industrial enzymes and could potentially be employed for other design goals.",
keywords = "Enzyme production levels, Protein redesign, Redesign by learning",
author = "{Van Den Berg}, {Bastiaan A.} and Reinders, {Marcel J T} and {Van Der Laan}, {Jan Metske} and J. Roubos and {De Ridder}, Dick",
year = "2014",
doi = "10.1093/protein/gzu031",
language = "English",
volume = "27",
pages = "281--288",
journal = "Protein Engineering, Design and Selection",
issn = "1741-0126",
publisher = "Oxford University Press",
number = "9",

}

TY - JOUR

T1 - Protein redesign by learning from data

AU - Van Den Berg, Bastiaan A.

AU - Reinders, Marcel J T

AU - Van Der Laan, Jan Metske

AU - Roubos, J.

AU - De Ridder, Dick

PY - 2014

Y1 - 2014

N2 - Protein redesign methods aim to improve a desired property by carefully selecting mutations in relevant regions guided by protein structure. However, often protein structural requirements underlying biological characteristics are not well understood. Here, we introduce a methodology that learns relevant mutations from a set of proteins that have the desired property and demonstrate it by successfully improving production levels of two enzymes by Aspergillus niger, a relevant host organism for industrial enzyme production. We validated our method on two enzymes, an esterase and an inulinase, creating four redesigns with 5-45 mutations. Up to 10-fold increase in production was obtained with preserved enzyme activity for small numbers of mutations, whereas production levels and activities dropped for too aggressive redesigns. Our results demonstrate the feasibility of protein redesign by learning. Such an approach has great potential for improving production levels of many industrial enzymes and could potentially be employed for other design goals.

AB - Protein redesign methods aim to improve a desired property by carefully selecting mutations in relevant regions guided by protein structure. However, often protein structural requirements underlying biological characteristics are not well understood. Here, we introduce a methodology that learns relevant mutations from a set of proteins that have the desired property and demonstrate it by successfully improving production levels of two enzymes by Aspergillus niger, a relevant host organism for industrial enzyme production. We validated our method on two enzymes, an esterase and an inulinase, creating four redesigns with 5-45 mutations. Up to 10-fold increase in production was obtained with preserved enzyme activity for small numbers of mutations, whereas production levels and activities dropped for too aggressive redesigns. Our results demonstrate the feasibility of protein redesign by learning. Such an approach has great potential for improving production levels of many industrial enzymes and could potentially be employed for other design goals.

KW - Enzyme production levels

KW - Protein redesign

KW - Redesign by learning

UR - http://www.scopus.com/inward/record.url?scp=84906830857&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906830857&partnerID=8YFLogxK

U2 - 10.1093/protein/gzu031

DO - 10.1093/protein/gzu031

M3 - Article

C2 - 25082898

AN - SCOPUS:84906830857

VL - 27

SP - 281

EP - 288

JO - Protein Engineering, Design and Selection

JF - Protein Engineering, Design and Selection

SN - 1741-0126

IS - 9

ER -