Machine Learning in Protein Design

 

Home Page
 
Group Members
 
Meeting Dates
Progress
 
Datapassword required
 
Contact

 

Figure : Tetrahymena GCN5


This project is funded by the
Multidisciplinary Research
Opportunities for Women program
of the CRA-W
.
For more information, please see:
http://www.cra.org/Activities/craw/index.php

and
http://www.cra.org/Activities/craw/mrow/

 

Questions Addressed
How the machine learning algorithms can be used to predict the amino acid substitution of a protein sequence so that it will obtain higher activity and stability?

  • What is machine learning?
    A field of artificial intelligence in which the algorithms were used to improve the quality of the subject by learning through past experience

  • How are machine learning algorithms used in our project?
    In vivo incorporation of non-natural amino acids can be used to improve protein stability.  However, there is a trade off; improved stability of the protein may lead to loss in activity.  One way to improve function is to employ machine-learning algorithms to identify the variants that improve activity.  With the aid of computer guidance, we plan to design a set of fluorinated variants for our target protein Tetrahymena GCN5 (tGCN5).  Using information from biochemical and structural data, we identified eleven residues to mutate.  One of the supervised machine learning algorithms, Linear Regression Analysis, will be used to identify the residues that have positive impact on the mutated sequences.  


MRO-W 2007-2008 Mid-Year Progress Report


MRO-W 2007-2008 Final Report