I want to see the change in binding ability of the enzyme with the substrate in presence of a probable drug molecule insilico. Where can I find reliable information to learn about the basics from the start?
That depends on what you are interested in.
- If you just want to understand how the concentrations of enzyme, substrate and inhibitor affect reaction velocities (that is, enzyme kinetics), you can do that with Monte-Carlo simulations. This is covered in text books on model building in biology. The statistical system R (https://cran.r-project.org/) has the ability to perform such studies.
- If, however, you are interested in how binding of substrates and inhibitors changes the molecular structure of the enzyme, then you have to perform molecular modelling. This is theoretically complicated and numerically expensive. A.R. Leach: Molecular Modelling, 2nd ed., Harlow et al. (Pearson) 2001 would be a text book that covers this. Be warned though: In my personal experience, even for very simple problems, the results of such simulations have only limited resemblance with reality. For example, the 3D-structure of bilirubin is known from X-ray crystallography (doi:10.1098/rspb.1978.0066). The simulation with Avogadro (http://avogadro.cc/) will not reproduce it.
In Silico analysis is all about computational biology. Like Engelbert said, it depends on what you want to do. For gene expression analysis, see the link.Article In silico gene expression analysis - An overview
For In Silico analysis of proteins, see the attached file. For the challenges you might encounter see the attached document. The databases could be found in the document attached.
Thank you!
So the situation is that I have the AA sequence of the enzyme, but a confirmed 3D structure is unavailable. So this means I would have to use a simulated version of the molecule? How do I obtain this?
That again depends on the situation.
- Predictions of secondary structure have been performed since the '70s, using statistical information on preference of amino acids for secondary structures (doi:10.1016/0022-2836(78)90297-8, doi:10.1006/jmbi.1993.1413). Prediction accuracy is about 70%. Tertiary structure is not predictable by such methods.
- If your protein has a homologue with solved structure, it may be possible to "thread" your molecule onto it. If the proteins are related closely enough, prediction can be fairly accurate.
- In theory, it should be possible to calculate the secondary and tertiary structure of a protein by molecular modelling, using first principles. Bond angles have optimal values, it requires energy to increase or decrease them. The same is true for bond lengths. Hydration of hydrophobic residues requires energy, etc. Thus, each conformation of a protein has a total overall energy. The conformation of minimal energy is the native state. However, computationally this is a NP-complete problem, computational effort increases with the faculty of the number of atoms in the molecule. Without guidance by NMR, X-ray crystallography or electron crystallography even small proteins are out of reach for our computers, even with distributed computing (folding.stanford.edu, boinc.bakerlab.org/rosetta/). I have discussed all this in ISBN 978-1-4419-7250-7, chapters 32-34.
- As mentioned above, even for small molecules the results of modelling are less than shining.
- Even the theory that the native state of a protein is the one with the lowest free energy is probably incorrect in some cases, as shown by amyloidoses.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs