---

---

Secret team meetings and the sharing of 3.2Terabytes of free software -->HERE!... Don't spy, we don't like spies!

At the moment, we are primarily working on evaluating, and improving, Rosetta.
So we select small proteins from the PDB that have been solved to high resolution.

At the moment, we are primarily working on evaluating, and improving, Rosetta.
So we select small proteins from the PDB that have been solved to high resolution.

I might start a new thread on this.. but what algorythm are we using in this?

I was having a bud lite moment the other day and was thinking, I bet bayesian analysis could be used and would be a powerful tool in rendering lowest energy level.

---

Secret team meetings and the sharing of 3.2Terabytes of free software -->HERE!... Don't spy, we don't like spies!

I might start a new thread on this.. but what algorythm are we using in this?

I was having a bud lite moment the other day and was thinking, I bet bayesian analysis could be used and would be a powerful tool in rendering lowest energy level.

Quick answer: we do use at least Bayes Rule in the creation of our energy function.

We want to know the probability of a folded structure, F, given an amino acid squence, S. Call this P( F | S ). By Bayes Rule we can calculate it from P(F) P(S|F) / P(S). These terms are easier to estimate that P(F|S) itself. P(F), the probability of a fold independent of sequence, can be estimated from PDB statistics. P(S|F) can also be estimated by examing the frequencies of different amino acids in different folded conformations. Our low resolution scoring function is built on these ideas. If you have access to scientific journals, you might look at the following paper: Simons, K. T., Ruczinski, I., Kooperberg, C., Fox, B. A., Bystroff, C., Baker, D. (1999). Improved recognition of native-like protein structures using a combination of sequence-dependent and sequence-independent features of proteins Proteins 34, 82-95.

Ill check out the paper, thanx..


Its amazing how much we use Bayes work at present.

I might start a new thread on this.. but what algorythm are we using in this?

I was having a bud lite moment the other day and was thinking, I bet bayesian analysis could be used and would be a powerful tool in rendering lowest energy level.

Quick answer: we do use at least Bayes Rule in the creation of our energy function.

We want to know the probability of a folded structure, F, given an amino acid squence, S. Call this P( F | S ). By Bayes Rule we can calculate it from P(F) P(S|F) / P(S). These terms are easier to estimate that P(F|S) itself. P(F), the probability of a fold independent of sequence, can be estimated from PDB statistics. P(S|F) can also be estimated by examing the frequencies of different amino acids in different folded conformations. Our low resolution scoring function is built on these ideas. If you have access to scientific journals, you might look at the following paper: Simons, K. T., Ruczinski, I., Kooperberg, C., Fox, B. A., Bystroff, C., Baker, D. (1999). Improved recognition of native-like protein structures using a combination of sequence-dependent and sequence-independent features of proteins Proteins 34, 82-95.

---

Secret team meetings and the sharing of 3.2Terabytes of free software -->HERE!... Don't spy, we don't like spies!