Excluded volume in IMP

There are many options for computing excluded volume like scores in IMP, each with advantages and disadvantages as well as different domains of applicability.

All the following produce a score that is the sum of the harmonic lower bound over all pairs of overlapping balls (minus ones that are filtered out or, when appropriate, excluded by the ClosePairsFinder used).

Method

advantages

disadvantages

IMP::container::PairsRestraint using an IMP::container::ClosePairContainer coupled with an IMP::core::HarmonicLowerBoundSphereDistancePairScore

This is the most flexible method as you can replace the pair score used with any other one, e.g. IMP::atom::DopePairScore or a Lennard-Jones term

can be slower than something that is less flexible

IMP::core::ExcludedVolumeRestraint

it should be faster (it is on benchmarks). If a maximum score is set and domino is used, it can be much faster as it can abort score evaluation when the maximum is exceeded

less flexible, the scoring function cannot be changed

IMP::container::PairsRestraint using an IMP::container::AllPairContainer and an IMP::core::HarmonicLowerBoundSphereDistancePairScore

this can be faster than the above scoring functions when there are relatively few particles involved as it is much simpler

always computes all pairs so slow when there are many pairs

Lots of IMP::core::PairRestraints using IMP::core::HarmonicLowerBoundSphereDistancePairScore

none, really (although, internally, most of the above are converted to this when DOMINO is used)

slow

IMP::container::PairsRestraint using an IMP::container::ClosePairContainer or IMP::core::PairRestraints on the rigid bodies coupled with an IMP::core::ClosePairsPairScore

not clear this ever makes sense

generally slower and more cumbersome to set up

The methods below produce slightly different scores, that may be more or less appropriate depending on the context.

IMP: excluded_volume (last edited 2011-11-02 23:45:48 by DanielRussel)