Machine learning to estimate the local quality of protein crystal structures.
Ontology highlight
ABSTRACT:
SUBMITTER: Miyaguchi I
PROVIDER: S-EPMC8654820 | biostudies-literature | 2021 Dec
REPOSITORIES: biostudies-literature
Publications
Machine learning to estimate the local quality of protein crystal structures.
Miyaguchi Ikuko I Sato Miwa M Kashima Akiko A Nakagawa Hiroyuki H Kokabu Yuichi Y Ma Biao B Matsumoto Shigeyuki S Tokuhisa Atsushi A Ohta Masateru M Ikeguchi Mitsunori M
Scientific reports 20211208 1
Low-resolution electron density maps can pose a major obstacle in the determination and use of protein structures. Herein, we describe a novel method, called quality assessment based on an electron density map (QAEmap), which evaluates local protein structures determined by X-ray crystallography and could be applied to correct structural errors using low-resolution maps. QAEmap uses a three-dimensional deep convolutional neural network with electron density maps and their corresponding coordinat ...[more]
Similar Datasets
Project description:Elucidation of mesoscopic structures of molecular systems is of considerable scientific and technological interest for the development and optimization of advanced materials. Molecular dynamics simulations are a promising means of revealing macroscopic physical properties of materials from a microscopic viewpoint, but analysis of the resulting complex mesoscopic structures from microscopic information is a non-trivial and challenging task. In this study, a Machine Learning-aided Local Structure Analyzer (ML-LSA) is developed to classify the complex local mesoscopic structures of molecules that have not only simple atomistic group units but also rigid anisotropic functional groups such as mesogens. The proposed ML-LSA is applied to classifying the local structures of liquid crystal polymer (LCP) systems, which are of considerable scientific and technological interest because of their potential for sensors and soft actuators. A machine learning (ML) model is constructed from small, and thus computationally less costly, monodomain LCP trajectories. The ML model can distinguish nematic- and smectic-like monodomain structures with high accuracy. The ML-LSA is applied to large, complex quenched LCP structures, and the complex local structures are successfully classified as either nematic- or smectic-like. Furthermore, the results of the ML-LSA suggest the best order parameter for distinguishing the two mesogenic structures. Our ML model enables automatic and systematic analysis of the mesogenic structures without prior knowledge, and thus can overcome the difficulty of manually determining the specific order parameter required for the classification of complex structures.
| S-EPMC6841663 | biostudies-literature
Project description:Crystallographic R work and R free values, which are measures of the ability of the models of macromolecular structures to explain the crystallographic data on which they are based, are often used to assess structure quality. It is widely known, and confirmed here that both are sensitive to the methods used to compute them, and can be manipulated to improve the apparent quality of the model. As an alternative it is proposed here that the quality of crystallographic models should be assessed using a global goodness-of-fit metric RO2A /R work where RO2A is the number of reflections used for refinement divided by the number of nonhydrogen atoms in the structure, and R work is the working R-factor of the refined structure. Also, analysis of structures in the Protein Data Bank suggests that many data sets have been truncated at high resolution, thereby improving the R-factor statistics. To discourage this practice, it is proposed that the resolution of a dataset be defined as the resolution of the shell of data where <I/σI > falls to 1. The proposed goodness-of-fit metric encourages investigators to use all the data available rather than a truncated subset.
| S-EPMC4420517 | biostudies-literature
Project description:Determining residue-level protein properties, such as sites of post-translational modifications (PTMs), is vital to understanding protein function. Experimental methods are costly and time-consuming, while traditional rule-based computational methods fail to annotate sites lacking substantial similarity. Machine Learning (ML) methods are becoming fundamental in annotating unknown proteins and their heterogeneous properties. We present ASAP (Amino-acid Sequence Annotation Prediction), a universal ML framework for predicting residue-level properties. ASAP extracts numerous features from raw sequences, and supports easy integration of external features such as secondary structure, solvent accessibility, intrinsically disorder or PSSM profiles. Features are then used to train ML classifiers. ASAP can create new classifiers within minutes for a variety of tasks, including PTM prediction (e.g. cleavage sites by convertase, phosphoserine modification). We present a detailed case study for ASAP: CleavePred, an ASAP-based model to predict protein precursor cleavage sites, with state-of-the-art results. Protein cleavage is a PTM shared by a wide variety of proteins sharing minimal sequence similarity. Current rule-based methods suffer from high false positive rates, making them suboptimal. The high performance of CleavePred makes it suitable for analyzing new proteomes at a genomic scale. The tool is attractive to protein design, mass spectrometry search engines and the discovery of new bioactive peptides from precursors. ASAP functions as a baseline approach for residue-level protein sequence prediction. CleavePred is freely accessible as a web-based application. Both ASAP and CleavePred are open-source with a flexible Python API.Database URL: ASAP's and CleavePred source code, webtool and tutorials are available at: https://github.com/ddofer/asap; http://protonet.cs.huji.ac.il/cleavepred.
| S-EPMC5045867 | biostudies-literature
Project description:Cryo-EM and X-ray crystallography provide crucial experimental data for obtaining atomic-detail models of biomacromolecules. Refining these models relies on library-based stereochemical restraints, which, in addition to being limited to known chemical entities, do not include meaningful noncovalent interactions relying solely on nonbonded repulsions. Quantum mechanical (QM) calculations could alleviate these issues but are too expensive for large molecules. We present a novel AI-enabled Quantum Refinement (AQuaRef) based on AIMNet2 neural network potential mimicking QM at substantially lower computational costs. By refining 41 cryo-EM and 30 X-ray structures, we show that this approach yields atomic models with superior geometric quality compared to standard techniques, while maintaining an equal or better fit to experimental data.
| S-EPMC11275739 | biostudies-literature
Project description:One fundamental problem of protein biochemistry is to predict protein structure from amino acid sequence. The inverse problem, predicting either entire sequences or individual mutations that are consistent with a given protein structure, has received much less attention even though it has important applications in both protein engineering and evolutionary biology. Here, we ask whether 3D convolutional neural networks (3D CNNs) can learn the local fitness landscape of protein structure to reliably predict either the wild-type amino acid or the consensus in a multiple sequence alignment from the local structural context surrounding site of interest. We find that the network can predict wild type with good accuracy, and that network confidence is a reliable measure of whether a given prediction is likely going to be correct or not. Predictions of consensus are less accurate and are primarily driven by whether or not the consensus matches the wild type. Our work suggests that high-confidence mis-predictions of the wild type may identify sites that are primed for mutation and likely targets for protein engineering.
| S-EPMC8603988 | biostudies-literature
Project description:Chemical stability is a major concern in the development of protein therapeutics due to its impact on both efficacy and safety. Protein "hotspots" are amino acid residues that are subject to various chemical modifications, including deamidation, isomerization, glycosylation, oxidation etc. A more accurate prediction method for potential hotspot residues would allow their elimination or reduction as early as possible in the drug discovery process. In this work, we focus on prediction models for asparagine (Asn) deamidation. Sequence-based prediction method simply identifies the NG motif (amino acid asparagine followed by a glycine) to be liable to deamidation. It still dominates deamidation evaluation process in most pharmaceutical setup due to its convenience. However, the simple sequence-based method is less accurate and often causes over-engineering a protein. We introduce structure-based prediction models by mining available experimental and structural data of deamidated proteins. Our training set contains 194 Asn residues from 25 proteins that all have available high-resolution crystal structures. Experimentally measured deamidation half-life of Asn in penta-peptides as well as 3D structure-based properties, such as solvent exposure, crystallographic B-factors, local secondary structure and dihedral angles etc., were used to train prediction models with several machine learning algorithms. The prediction tools were cross-validated as well as tested with an external test data set. The random forest model had high enrichment in ranking deamidated residues higher than non-deamidated residues while effectively eliminated false positive predictions. It is possible that such quantitative protein structure-function relationship tools can also be applied to other protein hotspot predictions. In addition, we extensively discussed metrics being used to evaluate the performance of predicting unbalanced data sets such as the deamidation case.
| S-EPMC5521779 | biostudies-literature
Project description:Using chemical laboratory procedures to estimate the fruit quality parameters (biochemical parameters) of mango "Succarri" and strawberry "Florida" as indicators of ripening degrees in a large area presents challenges such as low throughput, labor intensity, time consumption, and the need for multiple samples. So, using spectral reflectance-based proximal remote sensing to quickly and accurately measure biochemical parameters in different fruits is important to find the best time to harvest, make food ripen faster, and the processing of food easier. This has significant economic and ecological advantages. The objective of this study was to evaluate the biochemical parameters of mango and strawberry fruits at various ripening stages. This was done by utilizing a combination of established and newly developed spectral reflectance indices (SRIs) in conjunction with machine learning (ML) models, including artificial neural networks (ANN), random forests (RF), and decision trees (DT). For mango fruit, the parameters estimated were chlorophyll content, total soluble solids (TSS), and firmness, whereas for strawberry fruit, the parameters were L*, b*, TSS, and firmness. These results revealed significant differences in SRI values across various ripening stages, indicating variances in the fruit's biochemical parameters. The newly developed SRIs showed superior efficacy in evaluating these parameters. The integration of SRIs with diverse ML models proved to be a successful strategy for precisely estimating biochemical parameters. For mango's biochemical parameter prediction, the ANN models demonstrated R2 values ranging from 0.92 to 1.00 and from 0.93 to 0.98 for training and testing, respectively. On the other hand, the RF models exhibited R2 values ranging from 0.98 to 1.00 and from 0.93 to 0.99 during training and testing, respectively. The DT models showed high performance, with R2 values ranging from 0.95 to 1.00 and from 0.88 to 0.99 for the training and testing phases. For strawberry's biochemical parameter prediction, the ANN models achieved R2 values between 0.75 and 0.91 and between 0.58 and 0.91 during training and testing phases, respectively. On the other hand, RF models showed R2 values between 0.85 and 0.91 during training and between 0.74 and 0.86 during testing. The DT models demonstrated excellent results, with R2 values ranging from 0.75 to 0.91 for the training set and 0.74 to 0.81 for the testing set. It can be concluded that combining SRIs with ML models, such as ANN, RF, and DT, can accurately predict the biochemical properties of mango and strawberry fruits.
| S-EPMC11813159 | biostudies-literature
Project description:Protein structure prediction has long been available as an alternative to experimental structure determination, especially via homology modeling based on templates from related sequences. Recently, models based on distance restraints from coevolutionary analysis via machine learning to have significantly expanded the ability to predict structures for sequences without templates. One such method, AlphaFold, also performs well on sequences where templates are available but without using such information directly. Here we show that combining machine-learning based models from AlphaFold with state-of-the-art physics-based refinement via molecular dynamics simulations further improves predictions to outperform any other prediction method tested during the latest round of CASP. The resulting models have highly accurate global and local structures, including high accuracy at functionally important interface residues, and they are highly suitable as initial models for crystal structure determination via molecular replacement.
| S-EPMC7125014 | biostudies-literature
Project description:In-vivo estimation of mechanical properties of the myocardium is essential for patient-specific diagnosis and prognosis of cardiac disease involving myocardial remodeling, including myocardial infarction and heart failure with preserved ejection fraction. Current approaches use time-consuming finite-element (FE) inverse methods that involve reconstructing and meshing the heart geometry, imposing measured loading, and conducting computationally expensive iterative FE simulations. In this paper, we propose a machine learning (ML) model that feasibly and accurately predicts passive myocardial properties directly from select geometric, architectural, and hemodynamic measures, thus bypassing exhaustive steps commonly required in cardiac FE inverse problems. Geometric and fiber-orientation features were chosen to be readily obtainable from standard cardiac imaging protocols. The end-diastolic pressure-volume relationship (EDPVR), which can be obtained using a single-point pressure-volume measurement, was used as a hemodynamic (loading) feature. A comprehensive ML training dataset in the geometry-architecture-loading space was generated, including a wide variety of partially synthesized rodent heart geometry and myofiber helicity possibilities, and a broad range of EDPVRs obtained using forward FE simulations. Latin hypercube sampling was used to create 2500 examples for training, validation, and testing. A multi-layer feed-forward neural network (MFNN) was used as a deep learning agent to train the ML model. The model showed excellent performance in predicting stiffness parameters [Formula: see text] and [Formula: see text] associated with fiber direction ([Formula: see text] and [Formula: see text]). After conducting permutation feature importance analysis, the ML performance further improved for [Formula: see text] ([Formula: see text]), and the left ventricular volume and endocardial area were found to be the most critical geometric features for accurate predictions. The ML model predictions were evaluated further in two cases: (i) rat-specific stiffness data measured using ex-vivo mechanical testing, and (ii) patient-specific estimation using FE inverse modeling. Excellent agreements with ML predictions were found for both cases. The trained ML model offers a feasible technology to estimate patient-specific myocardial properties, thus, bridging the gap between EDPVR, as a confounded organ-level metric for tissue stiffness, and intrinsic tissue-level properties. These properties provide incremental information relative to traditional organ-level indices for cardiac function, improving the clinical assessment and prognosis of cardiac diseases.
| S-EPMC8971532 | biostudies-literature
Project description:Protein secondary structure discrimination is crucial for understanding their biological function. It is not generally possible to invert spectroscopic data to yield the structure. We present a machine learning protocol which uses two-dimensional UV (2DUV) spectra as pattern recognition descriptors, aiming at automated protein secondary structure determination from spectroscopic features. Accurate secondary structure recognition is obtained for homologous (97%) and nonhomologous (91%) protein segments, randomly selected from simulated model datasets. The advantage of 2DUV descriptors over one-dimensional linear absorption and circular dichroism spectra lies in the cross-peak information that reflects interactions between local regions of the protein. Thanks to their ultrafast (∼200 fs) nature, 2DUV measurements can be used in the future to probe conformational variations in the course of protein dynamics.
| S-EPMC9171355 | biostudies-literature