Biology subjects

Richardson, D. C.

Publications and source records attributed to Richardson, D. C..

5 recordsLinked to original sources

bioRxiv2025

Categorizing prediction modes within low-pLDDT regions of AlphaFold2 structures

AlphaFold2 protein structure predictions are widely available for structural biology uses. These predictions, especially for eukaryotic proteins, frequently contain extensive regions predicted below the pLDDT 70 level, the rule-of-thumb cutoff for high confidence. This work identifies major modes of behavior within low-pLDDT regions through a survey of human proteome predictions provided by the AlphaFold Protein Structure Database. The near-predictive mode resembles folded protein and can be a nearly accurate prediction. Barbed wire is extremely unproteinlike, being recognized by wide looping coils, an absence of packing contacts, and numerous signature validation outliers, and it likely represents a nonpredicted region. Pseudostructure presents an intermediate behavior with a misleading appearance of isolated and badly formed secondary structure-like elements. These prediction modes are compared with annotations of disorder from MobiDB, showing general correlation between barbed wire/pseudostructure and many measures of disorder, an association between pseudostructure and signal peptides, and an association between near-predictive and regions of conditional folding. To enable users to identify these regions within a prediction, a new Phenix tool is developed encompassing the results of this work, including prediction annotation, visual markup, and residue selection based on these prediction modes. This tool will help users develop expertise in interpreting difficult AlphaFold predictions and identify the near-predictive regions that can aid in molecular replacement when a prediction does not contain enough high-pLDDT regions.

biochemistry↗

bioRxiv2021

The importance of residue-level filtering, and the Top2018 best-parts dataset of high-quality protein residues

We have curated a high-quality, "best parts" reference dataset of about 3 million protein residues in about 15,000 PDB-format coordinate files, each containing only residues with good electron density support for a physically acceptable model conformation. The resulting pre-filtered data typically contains the entire core of each chain, in quite long continuous fragments. Each reference file is a single protein chain, and the total set of files were selected for low redundancy, high resolution, good MolProbity score, and other chain-level criteria. Then each residue was critically tested for adequate local map quality to firmly support its conformation, which must also be free of serious clashes or covalent-geometry outliers. The resulting Top2018 pre-filtered datasets have been released on the Zenodo online web service and is freely available for all uses under a Creative Commons license. Currently, one dataset is residue-filtered on mainchain plus C{beta} atoms, and a second dataset is full-residue filtered; each is available at 4 different sequence-identity levels. Here, we illustrate both statistics and examples that show the beneficial consequences of residue-level filtering. That process is necessary because even the best of structures contain a few highly disordered local regions with poor density and low-confidence conformations that should not be included in reference data. Therefore the open distribution of these very large, pre-filtered reference datasets constitutes a notable advance for structural bioinformatics and the fields that depend upon it. The Top2018 dataset provides the first representative sample of 3D protein structure for which excellence of experimental data constrains the detailed local conformation to be correct for essentially all 3 million residues included. Earlier generations of residue-filtered datasets were central in developing MolProbity validation used worldwide, and now Zenodo has enabled anyone to use out latest version as a sound basis for structural bioinformatics, protein design, prediction, improving biomedically important structures, or other applications.

biochemistry↗

bioRxiv2019

CaBLAM for CryoEM Chiropraxis, UnDowser to Rethink \"Waters\", and NGL Viewer to Recapture Online 3D Graphics in MolProbity Validation

The MolProbity web service provides macromolecular model validation to help correct local errors, for the structural biology community worldwide. Here we highlight new validation features, and also describe how we are fighting back against recent outside changes that degrade or endanger that mission. Sophisticated hacking of the MolProbity server has required continual monitoring and various security measures short of restricting user access. Refinement software now increasingly restrains validation criteria in order to supplement the sparser experimental data at 3-4[A] resolutions typical of modern cryoEM. But unfortunately the broad density allows optimization without fixing underlying problems, which means these structures often score much better than they really are. CaBLAM, our first new tool designed for this regime, was described in the previous Tools issue, and here we demonstrate its effectiveness in diagnosing local errors even when other validation outliers have been artificially removed. The deprecation of Java applets now prevents KiNG interactive online display of outliers on the 3D model during a MolProbity run, but that important functionality is now recaptured with a modified version of the Javascript NGL Viewer. Other new changes are more straightforwardly good. We are moving to the Neo4j database (graphical rather than relational), and will soon have cleaner as well as much larger reference datasets. In addition to several minor new features, we have developed a tool called UnDowser that analyzes the properties and context of modeled but clashing HOH \"waters\" to diagnose what they might actually represent. A dozen distinct scenarios are illustrated and described.

biophysics↗

bioRxiv2018

Measuring narrative engagement: The heart tells the story

Stories play a fundamental role in human culture. They provide a mechanism for sharing cultural identity, imparting knowledge, revealing beliefs, reinforcing social bonds and providing entertainment that is central to all human societies. Here we investigated the extent to which the delivery medium of a story (audio or visual) affected conscious and subconscious engagement with the narrative. Although participants self-reported greater involvement for watching video relative to listening to auditory scenes, stronger physiological responses were recorded for auditory stories. Sensors placed at their wrists showed higher and more variable heart rates, greater electrodermal activity, and even higher body temperatures. We interpret these findings as physiological evidence that the stories were more cognitively and emotionally engaging when presented in an auditory format. This may be because listening to a story, rather than watching a video, is a more active process of co-creation, and that this imaginative process in the listeners mind is detectable on the skin at their wrist.

neuroscience↗

bioRxiv2018

Cis-nonPro Peptides: Genuine Occurrences and their Functional Roles

While cis peptides preceding proline can occur about 5% of the time, cis peptides preceding any other residue ("cis-nonPro" peptides) are an extremely rare feature in protein structures, of considerable importance for two opposite reasons. On one hand, their genuine occurrences are mostly found at sites critical to biological function, from the active sites of carbohydrate enzymes to rare adjacent-residue disulfide bonds. On the other hand, a cis-nonPro can easily be misfit into weak or ambiguous electron density, which led to a high incidence of unjustified cis-nonPro over the 2006-2015 decade. This paper uses high-resolution crystallographic data and especially stringent quality-filtering at the residue level to identify genuine occurrences of cis-nonPro, and to survey both individual examples and broad patterns of their functionality. We explain the procedure developed to identify genuine cis-nonPro examples with almost no false positives. We then survey a large sample of the varied functional roles and structural contexts of cis-nonPro, including the uses of specific amino acids for particular purposes. We emphasize aspects not previously covered: that cis-nonPro always (except for vicinal disulfides) occur in highly ordered structure, and especially the great concentration of occurrence in proteins that process or bind carbohydrates (identified by occurrence on the CAZy website).

biochemistry↗