A5078589723- Computational Drug Discovery Methods
- Pulsars and Gravitational Waves Research
- Bioinformatics and Genomic Networks
- Geophysics and Gravity Measurements
- Machine Learning and Data Classification
- Venomous Animal Envenomation and Studies
- Pharmacogenetics and Drug Metabolism
- Evolutionary Algorithms and Applications
- Explainable Artificial Intelligence (XAI)
- Biochemical and Structural Characterization
- Radio Astronomy Observations and Technology
- Machine Learning in Bioinformatics
- Nicotinic Acetylcholine Receptors Study
- Biomedical Text Mining and Ontologies
- Cell Image Analysis Techniques
- Machine Learning in Materials Science
- Artificial Intelligence in Healthcare
- Genomics and Phylogenetic Studies
- Cosmology and Gravitation Theories
- Machine Learning in Healthcare
- Genetic Associations and Epidemiology
- Neural Networks and Applications
- Scientific Computing and Data Management
- Alzheimer's disease research and treatments
- Relativity and Gravitational Theory
University of Pennsylvania
2020-2024
Texas Tech University
2023-2024
California University of Pennsylvania
2024
The University of Texas Rio Grande Valley
2018-2023
Virginia Commonwealth University Medical Center
2022
Virginia Commonwealth University
2008-2021
Columbia University
2015-2019
Center for Genomic Science
2019
Weatherford College
2019
University of Vermont
2014-2018
Pulsar timing arrays (PTAs) use an array of millisecond pulsars to search for gravitational waves in the nanohertz regime pulse time arrival data. This paper presents rigorous tests PTA methods, examining their consistency across relevant parameter space. We discuss updates 15-year isotropic gravitational-wave background analyses and corresponding code representations. Descriptions internal structure flagship algorithms enterprise ptmcmcsampler are given facilitate understanding likelihood...
Novel machine learning and statistical modeling studies rely on standardized comparisons to existing methods using well-studied benchmark datasets. Few tools exist that provide rapid access many of these datasets through a standardized, user-friendly interface integrates well with popular data science workflows.
The Laser Interferometer Space Antenna (LISA) has two scientific objectives of cosmological focus: to probe the expansion rate universe, and understand stochastic gravitational-wave backgrounds their implications for early universe particle physics, from MeV Planck scale. However, range potential applications gravitational wave observations extends well beyond these objectives. This publication presents a summary state art in LISA cosmology, theory methods, identifies new opportunities use...
In drug development, a major reason for attrition is the lack of understanding cellular mechanisms governing toxicity. The black-box nature conventional classification models has limited their utility in identifying toxicity pathways. Here we developed DTox (deep learning toxicology), an interpretation framework knowledge-guided neural networks, which can predict compound response to assays and infer pathways individual compounds. We demonstrate that achieve same level predictive performance...
Pulsar timing array collaborations, such as the North American Nanohertz Observatory for Gravitational Waves (NANOGrav), are seeking to detect nanohertz gravitational waves emitted by supermassive black hole binaries formed in aftermath of galaxy mergers. We have searched continuous from individual circular using NANOGrav's recent 12.5-year data set. created new methods accurately model uncertainties on pulsar distances our analysis, and we implemented techniques account a common red noise...
Automated machine learning (AutoML) and deep (DL) are two cutting-edge paradigms used to solve a myriad of inductive tasks. In spite their successes, little guidance exists for when choose one approach over the other in context specific real-world problems. Furthermore, relatively few tools exist that allow integration both AutoML DL same analysis yield results combining strengths. Here, we seek address these issues, by (1.) providing head-to-head comparison binary classification on 6...
Background As global populations age and become susceptible to neurodegenerative illnesses, new therapies for Alzheimer disease (AD) are urgently needed. Existing data resources drug discovery repurposing fail capture relationships central the disease’s etiology response drugs. Objective We designed Alzheimer’s Knowledge Base (AlzKB) alleviate this need by providing a comprehensive knowledge representation of AD candidate therapeutics. Methods AlzKB as large, heterogeneous graph base...
A typical task in bioinformatics consists of identifying which features are associated with a target outcome interest and building predictive model. Automated machine learning (AutoML) systems such as the Tree-based Pipeline Optimization Tool (TPOT) constitute an appealing approach to this end. However, biomedical data, there often baseline characteristics subjects study or batch effects that need be adjusted for order better isolate on target. Thus, ability perform covariate adjustments...
ComptoxAI is a new data infrastructure for computational and artificial intelligence research in predictive toxicology. Here, we describe showcase ComptoxAI's graph-structured knowledge base the context of three real-world use-cases, demonstrating that it can rapidly answer complex questions about toxicology are infeasible using previous technologies resources. These use-cases each demonstrate tool information retrieval from being used to solve specific task: The "shortest path" module...
We answer frequently asked questions (FAQs) about the Hellings and Downs correlation curve -- "smoking-gun" signature that pulsar timing arrays (PTAs) have detected gravitational waves (GWs). Many of these arise from inadvertently applying intuition effects GWs on LIGO-like detectors to case timing, where not all it applies. This is because Earth-based detectors, like LIGO Virgo, arms are short (km scale) compared wavelengths they detect (approx 100-10,000 km). In contrast, PTAs respond...
<i>Background/Aims:</i> A previous study found a high prevalence of headaches in persons with familial Alzheimer’s disease (FAD) due to <i>PSEN1</i> mutation. In our we compared the between nondemented FAD mutation carriers (MCs) and non-mutation-carrying controls (NCs). <i>Methods:</i> headache questionnaire that assessed significant diagnosis migraine aura by ICHD-2 criteria was administered 27 individuals at risk for FAD. Frequency headaches, migraine,...
Pulsar timing arrays (PTAs) have made tremendous progress and are now showing strong evidence for the gravitational-wave background (GWB). Further probing origin characteristics of GWB will require more generalized analysis techniques. Bayesian methods most often used but can be computationally expensive. On other hand, frequentist methods, like PTA Optimal Statistic (OS), efficient produce results that complementary to allowing stronger statistical cases built from a confluence different...
Pulsar timing arrays (PTAs) detect gravitational waves (GWs) via the correlations they create in arrival times of pulses from different pulsars. The mean correlation, a function angle between directions to two pulsars, was predicted 1983 by Hellings and Downs (HD). Observation this angular pattern is ``smoking gun'' that GWs are present, so PTAs ``reconstruct HD curve'' estimating correlation using pulsar pairs separated similar angles. Several studies have examined amount which curve...
In a previous paper (gr-qc/0105100) we derived set of near-optimal signal detection techniques for gravitational wave detectors whose noise probability distributions contain non-Gaussian tails. The methods modify standard by truncating sample values which lie in those were derived, the frequentist framework, minimizing false alarm probabilities at fixed weak limit. For stochastic signals, resulting statistic consisted sum an auto-correlation term and cross-correlation term; it was necessary...
Animal venoms have been used for therapeutic purposes since the dawn of recorded history. Only a small fraction, however, tested pharmaceutical utility. Modern computational methods enable systematic exploration novel uses venom compounds. Unfortunately, there is currently no comprehensive resource describing clinical effects to support this analysis. We present VenomKB, new publicly accessible knowledge base and website that aims act as repository emerging putative therapies. Presently, it...
Venoms are a diverse and complex group of natural toxins that have been adapted to treat many types human disease, but rigorous computational approaches for discovering new therapeutic activities scarce. We designed validated platform—named VenomSeq—to systematically identify putative associations between venoms drugs/diseases via high-throughput transcriptomics perturbational differential gene expression analysis. In this study, we describe the architecture VenomSeq its evaluation using...
Abstract In computational toxicology, prediction of complex endpoints has always been challenging, as they often involve multiple distinct mechanisms. State‐of‐the‐art models are either limited by low accuracy, or lack interpretability due to their black‐box nature. Here, we introduce AIDTox, an interpretable deep learning model which incorporates curated knowledge chemical‐gene connections, gene‐pathway annotations, and pathway hierarchy. AIDTox accurately predicts cytotoxicity outcomes in...
Abstract Motivation Venom peptides comprise one of the richest sources bioactive compounds available for drug discovery. However, venom data and knowledge are fragmentary poorly structured, fail to capitalize on important characteristics venoms that make them so interesting biomedical community. Results We present VenomKB v2.0, a new open-access resource representation retrieval bioactivities, sequences, structures, classifications. provides complete infrastructure computational toxinology,...