Hi,
I'm doing directed evolution on one enzyme. the changes in amino acids affect the activity of the enzyme. from the analysis of mutants library, hydrophobicity and percentage helix are two dominant factors that influenced the activity.
I already done the DNA sequencing for 50 mutants out of 3000. From the sequences, I calculated the hydrophobicity and helix percentage. Now I need to develop a model to describe the relation between hydrophobicity and helix to the enzyme activity.
With only 50 mutants, I am not optimise to get a define model like RSM but I was hoping something similar to that. Where in the future, I may put in value of hydrophobicity and helix to predict the enzyme activity.
Thank you very much.
I don't know what RSM is. However, you have many different choices. Let's call your measure of enzymatic activity, A, the hydropathy index H and the percentage of helices as P. You may try linear regression models in the form A = c + beta1 H + beta2 P (eventually with an interaction term). First thing I would do is to check the relationship among H and A and P and A. are they (approx) linear? Also check the relationship among H and P. if they are correlated, you should not use them together in the regression.
Additional models can be imagined, but I think you can start with a linear approximation. Here you have 3 (or 4 with the interaction term) parameters therefore you should be able to estimate them from your data, provided the relationship is linear and your data contain the right information.
So your assumption is that the hydrophobicity (hp) and helix % (he) are the 2 independent variables, and the enzyme activity (ea) is the dependent variable? The very first step you should do is to make a 3D plot of z= f(x,y) where z is the ea, x as the hp and y as the he. That plot should suggest to you what next steps to take.
I suggest SAS software but it is hard to use if you are not familiar with.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology