Hi, I'm thinking of using the mbar overlap matrix to do some lambda scheduling related work, where the plan is to run a large number of lambda windows (e.g. 200 windows). Then based on the overlap matrix, choosing a subset of lambdas (e.g. 20 windows) to run the actual simulation.
I know that to infer the small mbar overlap matrix (20*20) from the big mbar overlap matrix (200*200) is not as simple as transferring the overlap between the corresponding lambda states from the big matrix into the small matrix (as they are normalised to sum up to one). And I wonder if I could get some advice on how to do this correctly and if there is already such functions in pymbar? Thanks.
Hi, I'm thinking of using the mbar overlap matrix to do some lambda scheduling related work, where the plan is to run a large number of lambda windows (e.g. 200 windows). Then based on the overlap matrix, choosing a subset of lambdas (e.g. 20 windows) to run the actual simulation.
I know that to infer the small mbar overlap matrix (
20*20) from the big mbar overlap matrix (200*200) is not as simple as transferring the overlap between the corresponding lambda states from the big matrix into the small matrix (as they are normalised to sum up to one). And I wonder if I could get some advice on how to do this correctly and if there is already such functions in pymbar? Thanks.