% IMPORTANT: The following is UTF-8 encoded.  This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.

@ARTICLE{Mulnaes:856174,
      author       = {Mulnaes, Daniel and Gohlke, Holger},
      title        = {{T}op{S}core: {U}sing {D}eep {N}eural {N}etworks and
                      {L}arge {D}iverse {D}ata {S}ets for {A}ccurate {P}rotein
                      {M}odel {Q}uality {A}ssessment},
      journal      = {Journal of chemical theory and computation},
      volume       = {14},
      number       = {11},
      issn         = {1549-9626},
      address      = {Washington, DC},
      reportid     = {FZJ-2018-05801},
      pages        = {6117–6126},
      year         = {2018},
      abstract     = {The value of protein models obtained with automated protein
                      structure prediction depends primarily on their accuracy.
                      Protein model quality assessment is thus critical to select
                      the model that can best answer biologically relevant
                      questions from an ensemble of predictions. However, despite
                      many advances in the field, different methods capture
                      different types of errors, begging the question of which
                      method to use. We introduce TopScore, a meta Model Quality
                      Assessment Program (meta-MQAP) that uses deep neural
                      networks to combine scores from 15 different primary
                      predictors to predict accurate residue-wise and
                      whole-protein error estimates. The predictions on six large
                      independent data sets are highly correlated to
                      superposition-independent errors in the model, achieving a
                      Pearson’s Rall2 of 0.93 and 0.78 for whole-protein and
                      residue-wise error predictions, respectively. This is a
                      significant improvement over any of the investigated primary
                      MQAPs, demonstrating that much can be gained by optimally
                      combining different methods and using different and very
                      large data sets.},
      cin          = {JSC / ICS-6 / NIC},
      ddc          = {610},
      cid          = {I:(DE-Juel1)JSC-20090406 / I:(DE-Juel1)ICS-6-20110106 /
                      I:(DE-Juel1)NIC-20090406},
      pnm          = {511 - Computational Science and Mathematical Methods
                      (POF3-511) / 551 - Functional Macromolecules and Complexes
                      (POF3-551) / Forschergruppe Gohlke $(hkf7_20170501)$},
      pid          = {G:(DE-HGF)POF3-511 / G:(DE-HGF)POF3-551 /
                      $G:(DE-Juel1)hkf7_20170501$},
      typ          = {PUB:(DE-HGF)16},
      pubmed       = {pmid:30252470},
      UT           = {WOS:000450695200057},
      doi          = {10.1021/acs.jctc.8b00690},
      url          = {https://juser.fz-juelich.de/record/856174},
}