Pairwise Segment Score?

Started by DoctorSockrates

DoctorSockrates Lv 1

I've never noticed this score category in the Segment Information score breakdown. Anyone got any knowledge on what this indicates?

rmoretti Staff Lv 1

The "Pairwise" component that you're seeing in recent puzzles is a measure of the electrostatic interaction between atoms. (The Rosetta term is fa_elec, for those who've read the Alford et al. paper.) For example, if you have a glutamate near to a lysine you'll likely get a good "Pairwise" score, and if you have two glutamates next to each other you'll probably get a bad score from it. (It's not just full charges, though. There's partial charges on each atom which can attract/repel)

These interactions were present prior to the recent score function update, but were put in the "Other" category. With the update they're better categorized. For the old hands which remember scoring before New Chapter, that had a "Pairwise" term (fa_pair in Rosetta), which captured much the same information, but in a more coarse-grain fashion.

jeff101 Lv 1

Can you give a more detailed reference for the Alford et al. paper,
like its year, title, and the journal in which it appeared?
If you can post a link to its pdf, that would be even more helpful.

Thanks!

Bruno Kestemont Lv 1

It's not Open access. There is a graphical summary. Here is the text abstract:

"Over the past decade, the Rosetta biomolecular modeling suite has informed diverse biological questions and engineering challenges ranging from interpretation of low-resolution structural data to design of nanomaterials, protein therapeutics, and vaccines. Central to Rosetta’s success is the energy function: a model parametrized from small-molecule and X-ray crystal structure data used to approximate the energy associated with each biomolecule conformation. This paper describes the mathematical models and physical concepts that underlie the latest Rosetta energy function, called the Rosetta Energy Function 2015 (REF15). Applying these concepts, we explain how to use Rosetta energies to identify and analyze the features of biomolecular models. Finally, we discuss the latest advances in the energy function that extend its capabilities from soluble proteins to also include membrane proteins, peptides containing noncanonical amino acids, small molecules, carbohydrates, nucleic acids, and other macromolecules."

LociOiling Lv 1

The first link works.

Now we need a Rosetta Stone for Rosetta, telling us where we can find "the repulsive van der Waals energy, fa_rep" and the rest in Foldit score parts.

(That van der Waals energy is sooo repulsive….)

rmoretti Staff Lv 1

Here's a "cheat sheet" for translating Rosetta terms (at least for the standard REF2015 terms) into Foldit score parts. Note that, depending on puzzle, there may be other terms which might show up in one of these categories or another one.

fa_atr - Packing

fa_rep - Clashing
fa_intra_rep - Clashing

fa_sol - Hiding
fa_intra_sol_xover4 - Hiding
lk_ball_wtd - Hiding

fa_elec - Pairwise

hbond_sr_bb - Bonding
hbond_lr_bb - Bonding
hbond_bb_sc - Bonding
hbond_sc - Bonding

dslf_fa13 - Disulfides

rama_prepro - Backbone
omega - Backbone
p_aa_pp - Backbone

fa_dun - Sidechain

cart_bonded - Ideality
yhh_planarity - Ideality

ref - Reference