Intrinsically Disordered Proteins, How Should I Proceed?

Question

[mudit.rustagi]

I am doing a project on intrinsically disordered proteins. Recently, I implemented a Support Vector Machine to classify proteins into structured or unstructured. My mentor now wants me to work on neurodegenerative disordered peptides, such as beta-amyloid, and tau (in the case of Alzheimer's). Which direction should I take? What kind of analyses should I be looking at (other than amino acid compostion)?

[Neilfws]

Cross-posted at Reddit.

Answer 1

[jackuser1979]

Intrinsically disordered protein has no commonly agreed definition. For implementing machine learning approach building data set is very important step.

For data set construction for protein disorder you can use
1. Disprot database - which is manually curated and frequently updated database for protein disorder
2. PDB: You can extract experimental structures which lack REMARK465 (regions lacking electron density in crystal structure) - see this reference

For implementing other than amino acid composition, you should read all the papers related to disorder prediction programmes and you have to think to implement either based on any of these methods or some new method which you have to think to get better predictor. see the list of predictors

For example

1. DISEMBL: which have three version of predictor- LOOPS (regions devoid of regular secondary structure); HOTLOOPS (highly mobile loops); REMARK465 (regions lacking electron density in crystal structure) which based on Neural networks trained on X-ray structure data.

2. DISOPRED2: Regions devoid of ordered regular secondary structure which based on Cascaded support vector machine classifiers trained on PSI-BLAST profiles.

3. DRIPRED2: Secondary structure based on Kohonen’s self organising map trained on sequences of known structure

4. Foldindex: Regions that have a low hydrophobicity and high net charge (either loops or unstructured regions)