William G. Noid

William G. Noid

Main Content

  • Associate Professor of Chemistry
528 Chemistry Building
University Park, PA 16802
(814) 867-2387


  1. B.S. University of Tennessee, Knoxville 2000.
  2. Ph.D. Cornell University, 2005.

Honors and Awards:

  1. Camille Dreyfus Teacher-Scholar Award 2012
  2. NSF Career Award 2011
  3. Alfred P. Sloan Foundation Research Fellow 2011
  4. Penn State Institute for CyberScience Faculty Fellow 2011
  5. HP Outstanding Junior Faculty Award from the ACS Division of Computers in Chemistry 2010
  6. NIH Ruth L. Kirschstein NRSA postdoctoral fellow 2006-2007
  7. Tunis Wentink Prize (co-recipient) 2005
  8. Wachter award in Theoretical/Physical Chemistry 2003
  9. NSF Graduate Research Fellow 2002-2005
  10. NSF IGERT Fellow in nonlinear dynamics and complex systems 2000-2002

Selected Publications:

Joseph F. Rudzinski and William G. Noid. “Bottom-up coarse-graining of peptide ensembles and helix-coil transitions.” J. Chem. Theor. Comput. 11 1278-91 (2015).

Chad W. Lawrence, Sushant Kumar, William G. Noid, and Scott A. Showalter. “The role of ordered proteins in the folding-upon-binding of intrinsically disordered proteins.” J. Phys. Chem. Lett. 5 833-8 (2014).

W. G. Noid “Perspective: Coarse-grained models for biomolecular systems.” J. Chem. Phys. 139 090901 (2013).

Christopher R. Ellis, Buddhadev Maiti, and W. G. Noid.  "Specific and Nonspecific Effects of Glycosylation."  J. Am. Chem. Soc. 134 8184-93 (2012).

Joseph F. Rudzinski and W. G. Noid "Coarse-graining entropy, forces, and structures." J. Chem. Phys. 135 214101 (2011).

J.W. Mullinax and W.G. Noid.  “Recovering physical potentials from a model protein databank.”  Proc. Natl. Acad. Sci. USA  107 (46) 19867-72. (2010).

J. W. Mullinax and W. G. Noid, “Generalized Yvon-Born-Green theory for molecular systems.” Phys. Rev. Lett. 103 198104 (2009).


Theories of statistical mechanics applied to investigate structural biology and materials science problems; theories for multiscale modeling; computational methods for coarse-grained modeling of liquids, peptides, and polymers; investigations of unfolded and intrinsically disordered proteins, as well as the impact of co-translational modifications.

Course-grained modeling

Despite remarkable advances in computational methods and resources, atomically detailed simulations remain prohibitively time-consuming for modeling many processes of interest, e.g., protein-protein interactions or self-assembly. These considerations have motivated tremendous interest in developing much more efficient “coarse-grained (CG) models.”  These CG models typically represent each molecule with relatively few interaction sites that each correspond to groups of one or more atoms.  By employing such a reduced representation of the system, CG models provide the computational efficiency that is necessary for addressing the physically relevant time- and length scales, as well as for providing adequate statistical sampling.  Furthermore, CG approaches also provide important conceptual advantages by eliminating “unnecessary” atomic details and focusing attention on the essential features of a system.  Consequently, CG models hold great promise for modeling many important “soft” materials, including liquids, surfactants, bilayers, polymers, proteins, nucleic acids, etc.

Of course, the utility of CG models fundamentally depends upon their ability to accurately describe the “correct physics” governing the phenomenon of interest.  However, it remains challenging to determine the appropriate interactions between the CG sites.  Moreover, a CG model that is parameterized to accurately model a particular protein, may not be “transferable” to other proteins, i.e., a CG model optimized for a particular protein may not be an accurate model for other proteins. 

Our research group derives and implements rigorous multiscale methods for addressing these challenges by developing accurate CG models on the basis of detailed simulations or, ultimately, experimental data.  Our recent generalization of the Yvon-Born-Green theory provides the first variational framework for directly (i.e., noniteratively) determining, from structural information alone, the set of potentials that provide an optimal approximation to the many-body potential of mean force for complex molecular systems.   We have introduced an extended ensemble framework as a variational theory for optimizing the transferability of CG potentials for modeling multiple systems.  We have established an information theoretic basis for this framework and have provided basic insight into the role of resolution for CG models.  Moreover, we have demonstrated the role of many-body correlations in determining CG potentials.  These advances direct our current work in attempting to realize the promise of multiscale modeling methods for complex systems. 

Disordered Proteins

Recent experimental results have generated renewed interest in the physical properties of partially unfolded proteins.  Although the conventional sequence-structure-function dogma suggests that biological function depends upon a well-defined three dimensional structure, considerable evidence now indicates that many “intrinsically disordered” proteins perform vital functions in critical cellular processes, such as transcription and translation.  Furthermore, partially unfolded proteins have been implicated in protein misfolding diseases, including debilitating neurodegenerative diseases such as Alzheimer’s Disease.  These results clearly motivate quantitative investigations of the conformational space relevant for intrinsically disordered and partially unfolded proteins. 

The dynamic and heterogeneous nature of unfolded protein ensembles present significant challenges for experimental techniques that report ensemble averaged measurements.  In close collaboration with leading experimental groups, such as the Showalter group at PSU, we are developing both theories and computational models for addressing this complexity.

Additionally, we have employed simulations to investigate the impact of N-linked glycosylation upon the biophysical properties of unfolded proteins.  Our simulation studies suggested that the biophysical effects of glycosylation are not simply due to steric interactions, but reflect specific attractive interactions, such as hydrogen bonding and hydrophobic interactions, between the peptide and glycan.  Moreover, our simulations suggest that these effects are very sensitive to the sidechain adjacent to the glycosylation site.  Finally, we demonstrated that the effects of glycosylation upon short peptides are surprisingly consistent with their effects upon full-size glycoproteins.

Research Interests:


Theories of statistical mechanics applied to investigate structural biology

Computational / Theoretical

Statistical mechanics of unfolded proteins


Theories of statistical mechanics applied to investigate structural biology

Document Actions