Overview

This page is where developers can collect benchmark results for various Rosetta protocols with beta_nov15 energy function. The main aim is to check any improvements or degradations in scientific performances before we switch our default energy function from talaris2014 to beta_nov15. Please make sure the items below are properly described when you contribute here.

Please refer to http://www.tablesgenerator.com/markdown_tables to make/edit tables. Also, if anyone has better idea on the format, please give a suggestion to Hahnbeom Park (hahnbeom@uw.edu).

Structure Prediction

Title Who Description Metric Talaris2014 BetaNov15 Sign? App/Mover Ref. Rosetta V.
Rotamer, core H Park CADRES Success rate + RTmin
Rotamer, interface H Park CADRES Success rate + RTmin
Decoy, Mike H Park CADRES Boltzmann P 0.538 0.600 + Relax, dual 1
Decoy, Patrick H Park CADRES Boltzmann P 0.606 0.699 + Relax, dual 1
PPdock, ZDOCK H Park CADRES Boltzmann P 0.712 0.779 + Relax, torsion 1
HomologyModeling H Park 67 CAMEO targets GDT-HA 63.9 65.1 + Hybridize 1
LoopModeling H Park Boltzmann P + NGK 2

Design

Title Who Description Metric Talaris2014 BetaNov15 Sign? App/Mover Ref. Rosetta V.
FullDesign H Park StdScientific SuccessRate 38.9 40.6 + fixbb 3
ProteinMono H Park One-res-at-a-time SuccessRate 45.1 47.0 + pilot app 1
ProteinProtein H Park One-res-at-a-time SuccessRate 45.1 47.0 + pilot app 1
ProteinLigand H Park One-res-at-a-time SuccessRate 45.1 47.0 + pilot app 1

Etc

Title Who Description Metric Talaris2014 BetaNov15 Sign? App/Mover Ref. Rosetta V.

References

  1. H Park et al, Simultaneous optimization of biomolecular energy function on features from small molecules and macromolecules, JCTC 2016.
  2. Stein et al,

See Also