The interconnection between brain regions in neurological disease encodes vital information for the advancement of biomarkers and diagnostics. Although graph convolutional networks are widely applied for discovering brain connection patterns that point to disease conditions, the potential of connection patterns that arise from multiple imaging modalities has yet to be fully realized. In this paper, we propose a multi-modal sparse interpretable GCN framework (SGCN) for the detection of Alzheimer's disease (AD) and its prodromal stage, known as mild cognitive impairment (MCI). In our experimentation, SGCN learned the sparse regional importance probability to find signature regions of interest (ROIs), and the connective importance probability to reveal disease-specific brain network connections. We evaluated SGCN on the Alzheimer's Disease Neuroimaging Initiative database with multi-modal brain images and demonstrated that the ROI features learned by SGCN were effective for enhancing AD status identification. The identified abnormalities were significantly correlated with AD-related clinical symptoms. We further interpreted the identified brain dysfunctions at the level of large-scale neural systems and sex-related connectivity abnormalities in AD/MCI. The salient ROIs and the prominent brain connectivity abnormalities interpreted by SGCN are considerably important for developing novel biomarkers. These findings contribute to a better understanding of the network-based disorder via multi-modal diagnosis and offer the potential for precision diagnostics. The source code is available at https://github.com/Houliang-Zhou/SGCN.

Video summarization techniques have been proven to improve the overall user experience when it comes to accessing and comprehending video content. If the user's preference is known, video summarization can identify significant information or relevant content from an input video, aiding them in obtaining the necessary information or determining their interest in watching the original video. Adapting video summarization to various types of video and user preferences requires significant training data and expensive human labeling. To facilitate such research, we proposed a new benchmark for video summarization that captures various user preferences. Also, we present a pipeline called Video Summarization with Language (VSL) for user-preferred video summarization that is based on pre-trained visual language models (VLMs) to avoid the need to train a video summarization system on a large training dataset. The pipeline takes both video and closed captioning as input and performs semantic analysis at the scene level by converting video frames into text. Subsequently, the user's genre preference was used as the basis for selecting the pertinent textual scenes. The experimental results demonstrate that our proposed pipeline outperforms current state-of-the-art unsupervised video summarization models. We show that our method is more adaptable across different datasets compared to supervised query-based video summarization models. In the end, the runtime analysis demonstrates that our pipeline is more suitable for practical use when scaling up the number of user preferences and videos.

Integrating brain images and genetic data provides a great opportunity to discover potential biomarkers for neurological disorder diagnosis. However, learning genetic information and brain network dysfunction remains a challenging task. In this paper, we propose an interpretable multi-modal imaging and genetic graph convolution network (GCN) for Alzheimer’s disease diagnosis. Our genetic network uses hierarchical GCN to mimic a gene ontology-based graph of biological processes and learn the information flow in this graph. In parallel, our imaging network uses a sparse interpretable GCN with node and edge importance probabilities to learn the brain network from multi-modal images. After multi-modal fusion, the final representation guided by a cluster-based consistency constraint is used to predict the disease-related clinical measures. We evaluate our method on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Our result shows that our imaging-genetics framework achieves superior prediction performance compared to all state-of-the-art methods. The interpretation demonstrated that the salient SNPs, and salient regions interpreted by important probabilities were significantly correlated with AD-related clinical symptoms, and considerably important for developing novel biomarkers. The code is available at https://github.com/Houliang-Zhou/IG-GCN.

High annotation costs serve as a significant hurdle in deploying modern deep learning architectures for clinically relevant medical applications, especially when dealing with the inherent heterogeneity of multimodal data, proving the critical need for innovative algorithms that can effectively utilize unlabeled data. In this paper, we propose a model named MCLCA, which integrates multimodal contrastive learning and cross-modal attention to diagnose Alzheimer’s Disease (AD) and identify biomarkers using both labeled and unlabeled multimodal brain imaging genetics data. Through multimodal contrastive learning, MCLCA can effectively learn representations even in the absence of sufficient labels. By utilizing cross-modal attention blocks, the model captures deep connections between different modalities, providing a more comprehensive view of diagnosis. Our proposed MCLCA model is evaluated using the ADNI database with three imaging modalities (VBM-MRI, FDG-PET, and AV45-PET) and genetic SNP data. The results demonstrate that MCLCA can identify important biomarkers with better prediction accuracy compared to the existing methods. The source code is available at https://github.com/MCLCA.

SUMMARY: Identifying genomic features responsible for genome-wide association study (GWAS) signals has proven to be a difficult challenge; many researchers have turned to colocalization analysis of GWAS signals with expression quantitative trait loci (eQTL) and splicing quantitative trait loci (sQTL) to connect GWAS signals to candidate causal genes. The ColocQuiaL pipeline provides a framework to perform these colocalization analyses at scale across the genome and returns summary files and locus visualization plots to allow for detailed review of the results. As an example, we used ColocQuiaL to perform colocalization between a recent type 2 diabetes GWAS and Genotype-Tissue Expression (GTEx) v8 single-tissue eQTL and sQTL data. AVAILABILITY AND IMPLEMENTATION: ColocQuiaL is primarily written in R and is freely available on GitHub: https://github.com/bvoightlab/ColocQuiaL.

We present a new three dimensional representation of hydrogen bond donors and acceptors as spherical cones. The conical representation describes the range of bond lengths and bond angles at which a hydrogen bond can form. We hypothesized that three dimensional intersections of these cones can predict the formation of hydrogen bonds and potentially their contribution to protein-protein interactions. As a result, this representation enables a new technique for identifying similarities in bond formation and bond geometry.

Many tools that explore models of protein complexes are also able to analyze interactions between specific residues and atoms. A comprehensive exploration of these interactions can often uncover aspects of protein-protein recognition that are not obvious using other protein analysis techniques. This paper describes DiffBond, a novel method for searching for intermolecular interactions between protein complexes while differentiating between three different types of interaction: hydrogen bonds, ionic bonds, and salt bridges. DiffBond incorporates textbook definitions of these three interactions while contending with uncertainties that are inherent in computational models of interacting proteins. We used it to examine the barnase-barstar, Rap1a-raf, and Smad2-Smad4 complexes, as well as a subset of protein complexes formed between three-finger toxins and nAChRs. Based on electrostatic interactions established by previous experimental studies, DiffBond was able to identify ionic and hydrogen bonds with high precision and recall, and identify salt bridges with high precision. In combination with other electrostatic analysis methods, DiffBond can be a useful tool in helping predict influential amino acids in protein-protein interactions and characterizing the type of interaction.

Comparing binding sites as geometric solids can reveal conserved features of protein structure that bind similar molecular fragments and varying features that select different partners. Due to the subtlety of these features, algorithmic efficiency and geometric precision are essential for comparison accuracy. For these reasons, this paper presents pClay, the first structure comparison algorithm to employ fine-grained parallelism to enhance both throughput and efficiency. We evaluated the parallel performance of pClay on both multicore workstation CPUs and a 61-core Xeon Phi, observing scaleable speedup in many thread configurations. Parallelism unlocked levels of precision that were not practical with existing methods. This precision has important applications, which we demonstrate: A statistical model of steric variations in binding cavities, trained with data at the level of precision typical of existing work, can overlook 46% of authentic steric influences on specificity (p <= .02). The same model, trained with more precise data from pClay, overlooked 0% using the same standard of statistical significance. These results demonstrate how enhanced efficiency and precision can advance the detection of binding mechanisms that influence specificity.

Algorithms for comparing protein structure are frequently used for function annotation. By searching for subtle similarities among very different proteins, these algorithms can identify remote homologs with similar biological functions. In contrast, few comparison algorithms focus on specificity annotation, where the identification of subtle differences among very similar proteins can assist in finding small structural variations that create differences in binding specificity. Few specificity annotation methods consider electrostatic fields, which play a critical role in molecular recognition. To fill this gap, this paper describes VASP-E (Volumetric Analysis of Surface Properties with Electrostatics), a novel volumetric comparison tool based on the electrostatic comparison of protein-ligand and protein-protein binding sites. VASP-E exploits the central observation that three dimensional solids can be used to fully represent and compare both electrostatic isopotentials and molecular surfaces. With this integrated representation, VASP-E is able to dissect the electrostatic environments of protein-ligand and protein-protein binding interfaces, identifying individual amino acids that have an electrostatic influence on binding specificity. VASP-E was used to examine a nonredundant subset of the serine and cysteine proteases as well as the barnase-barstar and Rap1a-raf complexes. Based on amino acids established by various experimental studies to have an electrostatic influence on binding specificity, VASP-E identified electrostatically influential amino acids with 100% precision and 83.3% recall. We also show that VASP-E can accurately classify closely related ligand binding cavities into groups with different binding preferences. These results suggest that VASP-E should prove a useful tool for the characterization of specific binding and the engineering of binding preferences in proteins.

BACKGROUND: Structural variations caused by a wide range of physico-chemical and biological sources directly influence the function of a protein. For enzymatic proteins, the structure and chemistry of the catalytic binding site residues can be loosely defined as a substructure of the protein. Comparative analysis of drug-receptor substructures across and within species has been used for lead evaluation. Substructure-level similarity between the binding sites of functionally similar proteins has also been used to identify instances of convergent evolution among proteins. In functionally homologous protein families, shared chemistry and geometry at catalytic sites provide a common, local point of comparison among proteins that may differ significantly at the sequence, fold, or domain topology levels. RESULTS: This paper describes two key results that can be used separately or in combination for protein function analysis. The Family-wise Analysis of SubStructural Templates (FASST) method uses all-against-all substructure comparison to determine Substructural Clusters (SCs). SCs characterize the binding site substructural variation within a protein family. In this paper we focus on examples of automatically determined SCs that can be linked to phylogenetic distance between family members, segregation by conformation, and organization by homology among convergent protein lineages. The Motif Ensemble Statistical Hypothesis (MESH) framework constructs a representative motif for each protein cluster among the SCs determined by FASST to build motif ensembles that are shown through a series of function prediction experiments to improve the function prediction power of existing motifs. CONCLUSIONS: FASST contributes a critical feedback and assessment step to existing binding site substructure identification methods and can be used for the thorough investigation of structure-function relationships. The application of MESH allows for an automated, statistically rigorous procedure for incorporating structural variation data into protein function prediction pipelines. Our work provides an unbiased, automated assessment of the structural variability of identified binding site substructures among protein structure families and a technique for exploring the relation of substructural variation to protein function. As available proteomic data continues to expand, the techniques proposed will be indispensable for the large-scale analysis and interpretation of structural data.

Many algorithms that compare protein structures can reveal similarities that suggest related biological functions, even at great evolutionary distances. Proteins with related function often exhibit differences in binding specificity, but few algorithms identify structural variations that effect specificity. To address this problem, we describe the Volumetric Analysis of Surface Properties (VASP), a novel volumetric analysis tool for the comparison of binding sites in aligned protein structures. VASP uses solid volumes to represent protein shape and the shape of surface cavities, clefts and tunnels that are defined with other methods. Our approach, inspired by techniques from constructive solid geometry, enables the isolation of volumetrically conserved and variable regions within three dimensionally superposed volumes. We applied VASP to compute a comparative volumetric analysis of the ligand binding sites formed by members of the steroidogenic acute regulatory protein (StAR)-related lipid transfer (START) domains and the serine proteases. Within both families, VASP isolated individual amino acids that create structural differences between ligand binding cavities that are known to influence differences in binding specificity. Also, VASP isolated cavity subregions that differ between ligand binding cavities which are essential for differences in binding specificity. As such, VASP should prove a valuable tool in the study of protein-ligand binding specificity.

BACKGROUND: Structural genomics projects such as the Protein Structure Initiative (PSI) yield many new structures, but often these have no known molecular functions. One approach to recover this information is to use 3D templates - structure-function motifs that consist of a few functionally critical amino acids and may suggest functional similarity when geometrically matched to other structures. Since experimentally determined functional sites are not common enough to define 3D templates on a large scale, this work tests a computational strategy to select relevant residues for 3D templates. RESULTS: Based on evolutionary information and heuristics, an Evolutionary Trace Annotation (ETA) pipeline built templates for 98 enzymes, half taken from the PSI, and sought matches in a non-redundant structure database. On average each template matched 2.7 distinct proteins, of which 2.0 share the first three Enzyme Commission digits as the template's enzyme of origin. In many cases (61%) a single most likely function could be predicted as the annotation with the most matches, and in these cases such a plurality vote identified the correct function with 87% accuracy. ETA was also found to be complementary to sequence homology-based annotations. When matches are required to both geometrically match the 3D template and to be sequence homologs found by BLAST or PSI-BLAST, the annotation accuracy is greater than either method alone, especially in the region of lower sequence identity where homology-based annotations are least reliable. CONCLUSION: These data suggest that knowledge of evolutionarily important residues improves functional annotation among distant enzyme homologs. Since, unlike other 3D template approaches, the ETA method bypasses the need for experimental knowledge of the catalytic mechanism, it should prove a useful, large scale, and general adjunct to combine with other methods to decipher protein function in the structural proteome.

This paper shows how to effectively combine a sampling-based method primarily designed for multiple-query motion planning [probabilistic roadmap method (PRM)] with sampling-based tree methods primarily designed for single-query motion planning (expansive space trees, rapidly exploring random trees, and others) in a novel planning framework that can be efficiently parallelized. Our planner not only achieves a smooth spectrum between multiple-query and single-query planning, but it combines advantages of both. We present experiments which show that our planner is capable of solving problems that cannot be addressed efficiently with PRM or single-query planners. A key advantage of our planner is that it is significantly more decoupled than PRM and sampling-based tree planners. Exploiting this property, we designed and implemented a parallel version of our planner. Our experiments show that our planner distributes well and can easily solve high-dimensional problems that exhaust resources available to single machines and cannot be addressed with existing planners.

We propose a combination of techniques that solve multiple queries for motion planning problems with single query planners. Our implementation uses a probabilistic roadmap method (PRM) with bidirectional rapidly exploring random trees (BI-RRT) as the local planner. With small modifications to the standard algorithms, we obtain a multiple query planner, which is significantly faster and more reliable than its component parts. Our method provides a smooth spectrum between the PRM and BI-RRT techniques and obtains the advantages of both. We observed that the performance differences are most notable in planning instances with several rigid nonconvex robots in a scene with narrow passages. Our work is in the spirit of non-uniform sampling and refinement techniques used in earlier work on PRM.