RosettaHoles: Rapid assessment of protein core packing for structure prediction, refinement, design, and validation

Sheffler, Will; Baker, David

doi:10.1002/pro.8

Cited by 141 publications

(145 citation statements)

References 18 publications

Supporting

Mentioning

140

Contrasting

Order By: Relevance

“…1, top panel) includes hydrophobic side chain atoms CH 1 , CH 2 , CH 3 as well as hydrophobic backbone atoms in loops, which are more similar to side chain hydrophobic atoms than those atoms in sheet or helix configurations. For the parameters up to the radius of hydrogen, this group of atoms displays spatial preferences consistent with more uniform packing in crystal structures and clumped arrangements in decoys, as observed previously 4 ; these hydrophobic atoms prefer a small amount of void volume immediately surrounding the atom and discourage void volume further from the atomic surface. This effect is observed for all atom types but is most pronounced for hydrophobic atoms.…”

supporting

confidence: 81%

“…However, the total cavity volume is not a good discriminator between computational predictions and corresponding X-ray structures (Supporting Information 2). The original version of RosettaHoles 4 showed that discrimination between computational models and crystal structures can be achieved using volumetric information. In this updated version, we seek to achieve better performance with a simplified method.…”

mentioning

confidence: 99%

“…Although this analysis was conducted with the Rosetta full-atom force field, in previous work we have observed that similar packing flaws appear in all computationally generated structures submitted to past Critical Assessment of Structure Prediction experiments. 4 For this reason, we would expect including RosettaHoles2 in other force fields to yield similar improvements in packing quality.…”

mentioning

confidence: 99%

See 2 more Smart Citations

RosettaHoles2: A volumetric packing measure for protein structure refinement and validation

Sheffler

Baker

2010

Protein Science

Self Cite

View full text Add to dashboard Cite

We present an improved version of RosettaHoles, a methodology for quantitative and visual characterization of protein core packing. RosettaHoles2 features a packing measure more rapidly computable, accurate and physically transparent, as well as a new validation score intended for structures submitted to the Protein Data Bank. The differential packing measure is parameterized to maximize the gap between computationally generated and experimentally determined X-ray structures, and can be used in refinement of protein structure models. The parameters of the model provide insight into components missing in current force fields, and the validation score gives an upper bound on the X-ray resolution of Protein Data Bank structures; a crystal structure should have a validation score as good as or better than its resolution.Keywords: protein folding; protein design; protein core packing; protein structure validation; crystallography Computationally generated protein structures are often incorrect, suggesting some aspect of the physical chemistry is not modeled properly. For Rosetta 1 models, bond geometry is ideal by construction, low scoring models have very few clashes, and van der waals attractive interactions are comparable to corresponding crystal structures; these models easily pass standard structure validation tests such as MolProbity 2 and Whatcheck. 3 In contrast, visual inspection suggests that computational models have more volumetric packing flaws than observed in crystal structures. This is at least in part due to some missing piece in the Rosetta energy function; crystal structures that are energy minimized in the Rosetta force field display the same packing flaws seen in decoy structures, though to a lesser extent (Supporting Information 1). A missing piece of the puzzle could be the cavity free energy, which is inherently volumetric in nature. However, the total cavity volume is not a good discriminator between computational predictions and corresponding X-ray structures (Supporting Information 2). The original version of RosettaHoles 4 showed that discrimination between computational models and crystal structures can be achieved using volumetric information. In this updated version, we seek to achieve better performance with a simplified method.To define a packing score that captures cavity free energy, we take an approach inspired by the implicit solvation model of Lazaridis and Karplus. 5

show abstract

supporting

confidence: 81%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

RosettaHoles2: A volumetric packing measure for protein structure refinement and validation

Sheffler

Baker

2010

Protein Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Each structure was subjected to relaxation under the Rosetta full-atom potential (21,22). Packing quality was analyzed using the RosettaHoles metric (16,17). A count of all buried, unsatisfied hydrogen bonds was obtained using a SASA probe of 1.4 Å. Hydrogen bonds were defined with a strict geometric criteria (23).…”

Section: Methodsmentioning

confidence: 99%

“…In the following, we present an automated protocol, Rosetta VIP (void identification and packing), for improving the packing of a structural model. We use the RosettaHoles analysis tool (16,17) and a simple geometric scoring function to identify a small set of mutations that may yield improved packing. We are able to devote more effort toward evaluating this reduced set of possibilities, including consideration of computationally expensive refinement steps incorporating backbone flexibility and off-rotamer side-chain freedom.…”

mentioning

confidence: 99%

Automated selection of stabilizing mutations in designed and natural proteins

Borgo

Havranek

2012

Proc. Natl. Acad. Sci. U.S.A.

120

View full text Add to dashboard Cite

The ability to engineer novel protein folds, conformations, and enzymatic activities offers enormous potential for the development of new protein therapeutics and biocatalysts. However, many de novo and redesigned proteins exhibit poor hydrophobic packing in their predicted structures, leading to instability or insolubility. The general utility of rational, structure-based design would greatly benefit from an improved ability to generate well-packed conformations. Here we present an automated protocol within the RosettaDesign framework that can identify and improve poorly packed protein cores by selecting a series of stabilizing point mutations. We apply our method to previously characterized designed proteins that exhibited a decrease in stability after a full computational redesign. We further demonstrate the ability of our method to improve the thermostability of a well-behaved native protein. In each instance, biophysical characterization reveals that we were able to stabilize the original proteins against chemical and thermal denaturation. We believe our method will be a valuable tool for both improving upon designed proteins and conferring increased stability upon native proteins.W ell-packed hydrophobic cores are a hallmark of protein structure (1, 2). Consistent with the central role the hydrophobic effect plays in protein stability, defects in core packing are associated with decreased stability (3) and loss of conformational specificity (4, 5). Analytical tools for the assessment of core packing are useful for identifying errors in experimentally determined structures and for rationalizing and predicting the effects of mutations on protein stability. Not surprisingly, the assembly of well-packed cores has been a central goal for computational protein design since its inception (6).Despite this central focus, models generated by computational protein design algorithms often exhibit poor hydrophobic packing. This is due to both the simplified structural representation and the scoring methods used by standard design protocols. Structurally, the protein backbone is treated as rigid, and different side-chain orientations are sampled at each position. These side chains are limited to a discrete set of commonly observed conformations known as rotamers. This combination implies a limited ability to fill arbitrary volumes compactly. The coarseness of this representation is exacerbated by the scoring functions, which generally include some form of the Lennard-Jones atomatom interaction term taken from molecular mechanics. There is a severe and well-known mismatch between the distances over which this term can vary strongly and the resolution afforded by a rotameric representation of side chains. Xiang and Honig addressed this problem by increasing the number of allowable rotamers until the scoring function could be satisfactorily sampled (7). More commonly, the scoring function is modified to accommodate the rotameric representation. This can be accomplished by reducing the atomic radii to mitigate clashes or by...

show abstract

Importance of Structure and Dynamics in the Rational Drug Design of G Protein‐Coupled Receptor ( GPCR ) Modulators

Lazim¹,

Lee²,

Nakliang³

et al. 2022

GPCRs as Therapeutic Targets

View full text Add to dashboard Cite

RosettaHoles: Rapid assessment of protein core packing for structure prediction, refinement, design, and validation

Cited by 141 publications

References 18 publications

RosettaHoles2: A volumetric packing measure for protein structure refinement and validation

RosettaHoles2: A volumetric packing measure for protein structure refinement and validation

Automated selection of stabilizing mutations in designed and natural proteins

Importance of Structure and Dynamics in the Rational Drug Design of G Protein‐Coupled Receptor ( GPCR ) Modulators

Contact Info

Product

Resources

About