Design of metalloproteins and novel protein folds
using variational autoencoders
- Submitting institution
-
University College London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 14521
- Type
- D - Journal article
- DOI
-
10.1038/s41598-018-34533-1
- Title of journal
- Scientific Reports
- Article number
- ARTN 16189
- First page
- -
- Volume
- 8
- Issue
- 1
- ISSN
- 2045-2322
- Open access status
- Compliant
- Month of publication
- November
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
2
- Research group(s)
-
-
- Citation count
- 13
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This is the first attempt to express the problem of designing proteins with both specified tertiary structure and with specified molecular functions, using a purely machine learning-based approach (Variational Autoencoders). Although VAEs had been used to model mutations in proteins before, our algorithm for encoding both a systematic representation of protein fold space and the molecular function was novel and has now been referenced in a numerous follow-up studies on using machine learning for protein design, including some recent ones reporting actual experimental validation of ML designed protein sequences, showing that these algorithms work both in silico and in vitro.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -