A5078026787- Protein Structure and Dynamics
- Enzyme Structure and Function
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- RNA and protein synthesis mechanisms
- Machine Learning in Materials Science
- Microbial Metabolic Engineering and Bioproduction
- Fractal and DNA sequence analysis
- Biomedical Text Mining and Ontologies
- Circadian rhythm and melatonin
- Bioinformatics and Genomic Networks
- Sleep and related disorders
- Advanced Electron Microscopy Techniques and Applications
- Software Engineering Research
- Glycosylation and Glycoproteins Research
- Sleep and Wakefulness Research
- Diet and metabolism studies
- Data Mining Algorithms and Applications
- Force Microscopy Techniques and Applications
- Algorithms and Data Compression
- Cancer-related gene regulation
- Topic Modeling
- Cardiovascular Health and Risk Factors
- Biochemical and Structural Characterization
- Epigenetics and DNA Methylation
University of Alabama at Birmingham
2025
University of Missouri
2020-2023
Samsung (India)
2012-2017
Sewanee: The University of the South
1989
Abstract We present the results for CAPRI Round 54, 5th joint CASP‐CAPRI protein assembly prediction challenge. The offered 37 targets, including 14 homodimers, 3 homo‐trimers, 13 heterodimers antibody–antigen complexes, and 7 large assemblies. On average ~70 CASP predictor groups, more than 20 automatics servers, submitted models each target. A total of 21 941 by these groups 15 scorer were evaluated using model quality measures DockQ score consolidating measures. performance was quantified...
Abstract We present the results for CAPRI Round 50, fourth joint CASP‐CAPRI protein assembly prediction challenge. The comprised a total of twelve targets, including six dimers, three trimers, and higher‐order oligomers. Four these were easy which good structural templates available either full assembly, or main interfaces (of oligomers). Eight difficult targets only distantly related found individual subunits. Twenty‐five groups eight automatic servers submitted ~1250 models per target....
Abstract To enhance the AlphaFold-Multimer-based protein complex structure prediction, we developed a quaternary prediction system (MULTICOM) to improve input fed AlphaFold-Multimer and evaluate refine its outputs. MULTICOM samples diverse multiple sequence alignments (MSAs) templates for generate structural predictions by using both traditional Foldseek-based alignments, ranks through complementary metrics, refines via Foldseek alignment-based refinement method. The with different...
Deep learning has revolutionized protein tertiary structure prediction recently. The cutting-edge deep methods such as AlphaFold can predict high-accuracy structures for most individual chains. However, the accuracy of predicting quaternary complexes consisting multiple chains is still relatively low due to lack advanced in field. Because interchain residue-residue contacts be used distance restraints guide modeling, here we develop a dilated convolutional residual network method (DRCon)...
Since the 14th Critical Assessment of Techniques for Protein Structure Prediction (CASP14), AlphaFold2 has become standard method protein tertiary structure prediction. One remaining challenge is to further improve its We developed a new version MULTICOM system sample diverse multiple sequence alignments (MSAs) and structural templates input generate models. The models are then ranked by both pairwise model similarity self-reported quality score. top refined novel alignment-based refinement...
Estimating the accuracy of quaternary structural models protein complexes and assemblies (EMA) is important for predicting structures applying them to studying function interaction. The pairwise similarity between proven useful estimating quality tertiary models, but it has been rarely applied models. Moreover, approach often fails when many are low similar each other. To address gap, we developed a hybrid method (MULTICOM_qa) combining score (PSS) an interface contact probability (ICPS)...
AlphaFold-Multimer has emerged as the state-of-the-art tool for predicting quaternary structure of protein complexes (assemblies or multimers) since its release in 2021. To further enhance AlphaFold-Multimer-based complex prediction, we developed a new prediction system (MULTICOM) to improve input fed and evaluate refine outputs generated by AlphaFold2-Multimer. Specifically, MULTICOM samples diverse multiple sequence alignments (MSAs) templates generate structural models using both traditional
Abstract Deep learning methods that achieved great success in predicting intrachain residue-residue contacts have been applied to predict interchain between proteins. However, these require multiple sequence alignments (MSAs) of a pair interacting proteins (dimers) as input, which are often difficult obtain because there not many known protein complexes available generate MSAs sufficient depth for In recognizing monomer forms homomultimers contain the co-evolutionary signals both and residue...
Proteins interact to form complexes. Predicting the quaternary structure of protein complexes is useful for function analysis, engineering, and drug design. However, few user-friendly tools leveraging latest deep learning technology inter-chain contact prediction distance-based modelling predict structures are available. To address this gap, we develop DeepComplex, a web server predicting dimeric It uses contacts in homodimer or heterodimer. The predicted then used construct dimer by...
We present the results for CAPRI Round 54, 5th joint CASP-CAPRI protein assembly prediction challenge. The offered 37 targets, including 14 homo-dimers, 3 homo-trimers, 13 hetero-dimers antibody-antigen complexes, and 7 large assemblies. On average ~70 CASP predictor groups, more than 20 automatics servers, submitted models each target. A total of 21941 by these groups 15 scorer were evaluated using model quality measures DockQ score consolidating measures. performance was quantified a...
Abstract Since CASP14, AlphaFold2 has become the standard method for protein tertiary structure prediction. One remaining challenge in field is to further improve accuracy of AlphaFold2-based To address this challenge, we developed a new version MULTICOM system sample diverse multiple sequence alignments (MSAs) and structural templates input generate models. The models are then ranked by both pairwise model similarity self-reported quality score. top refined novel alignment-based refinement...
Predicting the quaternary structure of protein complex is an important problem. Inter-chain residue-residue contact prediction can provide useful information to guide ab initio reconstruction structures. However, few methods have been developed build structures from predicted inter-chain contacts. Here, we develop first method based on gradient descent optimization (GD) dimers utilizing contacts as distance restraints. We evaluate GD several datasets homodimers and heterodimers using...
Computational biology is one of many scientific disciplines ripe for innovation and acceleration with the advent high-performance computing (HPC). In recent years, field machine learning has also seen significant benefits from adopting HPC practices. this work, we present a novel pipeline that incorporates various machine-learning approaches structure-based functional annotation proteins on scale whole genomes. Our makes extensive use deep provides computational insights into best practices...
Estimating the accuracy of quaternary structural models protein complexes and assemblies (EMA) is important for predicting structures applying them to studying function interaction. The pairwise similarity between proven useful estimating quality tertiary models, but it has been rarely applied models. Moreover, approach often fails when many are low similar each other. To address gap, we developed a hybrid method (MULTICOM_qa) combining score (PSS) an interface contact probability (ICPS)...
Abstract AlphaFold-Multimer has emerged as the state-of-the-art tool for predicting quaternary structure of protein complexes (assemblies or multimers) since its release in 2021. To further enhance AlphaFold-Multimer-based complex prediction, we developed a new prediction system (MULTICOM) to improve input fed and evaluate refine outputs generated by AlphaFold2-Multimer. Specifically, MULTICOM samples diverse multiple sequence alignments (MSAs) templates generate structural models using both...
Abstract Deep learning methods that achieved great success in predicting intrachain residue-residue contacts have been applied to predict interchain between proteins. However, these require multiple sequence alignments (MSAs) of a pair interacting proteins (dimers) as input, which are often difficult obtain because there not many known protein complexes available generate MSAs sufficient depth for In recognizing monomer forms homomultimers contain the co-evolutionary signals both and residue...
Abstract Motivation Deep learning has revolutionized protein tertiary structure prediction recently. The cutting-edge deep methods such as AlphaFold can predict high-accuracy structures for most individual chains. However, the accuracy of predicting quaternary complexes consisting multiple chains is still relatively low due to lack advanced in field. Because interchain residue-residue contacts be used distance restraints guide modeling, here we develop a dilated convolutional residual...
Environment plays a vital role in the sleep mechanism of human. It has been shown from many studies that sleeping and waking environment, time hours is very significant importance [1] which can result disorders variety diseases. This paper finds cycle an individual according changes ambient temperature to maximize his/her efficiency. We suggest method will assist increasing Using Fast-Fourier-Transformation (FFT) heart rate signals extract variability data such low frequency / high (LF/HF)...
Predicting the quaternary structure of protein complex is an important problem. Inter-chain residue-residue contact prediction can provide useful information to guide ab initio reconstruction structures. However, few methods have been developed build structures from predicted inter-chain contacts. Here, we introduce a gradient descent optimization algorithm (GD) dimers utilizing contacts as distance restraints. We evaluate GD on several datasets homodimers and heterodimers using true or...
In this paper, we introduce a Concept Graph Engine (CG-Engine) that generates patient specific personalized disease ranking based on the laboratory test data. CG-Engine uses Unified Medical Language System database as medical knowledge base. The consists of two concepts namely, concept graph and its attributes. is level tree starts at root node ends node. attributes are: Relation types, Semantic Number Sources Symmetric Information between nodes. These are used to compute weight tests...
Abstract Since demonstrating its outstanding performance in the 14th Critical Assessment of Techniques for Protein Structure Prediction (CASP14) 2020, AlphaFold2 has become standard method protein tertiary structure prediction. One remaining challenge field is to further improve accuracy AlphaFold2-based To address this challenge, we developed a new version MULTICOM system sample diverse multiple sequence alignments (MSAs) and structural templates input generate models. The models are then...
The authors propose a fast, area-efficient, bit-parallel systolic architecture for two's complement multiplication, which can be implemented on single VLSI chip. This is more efficient than other reported architectures in terms of performance and utilization silicon area. bit cells are arranged 2-D array enhances the extensibility provides efficiency high-precision data. Barrel shifters employed cell level instead basic full adder, increases throughput rate by factor three over pipeline...
A novel concept of Disease Based Temporal Score (DT-Score) is introduced to efficiently represent periodic laboratory test data. Through this score, temporal state an organ can be represented by summarizing The score used indicate early trend for chronic abnormalities and thus results in effective wellness measure. Many these have a late manifestation are major contributors healthcare cost mortality. Resources Unified Medical Language System (UMLS) automatic generation relational tree...