A5101786641- Research Data Management Practices
- Advanced Graph Neural Networks
- Semantic Web and Ontologies
- Retinal Imaging and Analysis
- Big Data and Business Intelligence
- Data Visualization and Analytics
- Biomedical Text Mining and Ontologies
- Advanced Chemical Sensor Technologies
- Cancer, Hypoxia, and Metabolism
- Machine Learning in Materials Science
- Biochemical Analysis and Sensing Techniques
- Protein Structure and Dynamics
- Olfactory and Sensory Function Studies
- Genetics, Bioinformatics, and Biomedical Research
- Glaucoma and retinal disorders
- Retinal Development and Disorders
- Biochemical and Molecular Research
- Enzyme Structure and Function
- Explainable Artificial Intelligence (XAI)
- Dementia and Cognitive Impairment Research
- Traumatic Brain Injury Research
- Digital Imaging for Blood Diseases
- Graph Theory and Algorithms
- Genetic Associations and Epidemiology
- Machine Learning in Bioinformatics
23andMe (United States)
2009-2023
Yale University
2022
New York University
2018-2021
Indiana University School of Medicine
2020
University of California, Irvine
2018
Massachusetts General Hospital
2010
Harvard University
2010
The Barbara Ann Karmanos Cancer Institute
2010
Wayne State University
2010
Stanford University
2006-2009
Formally published papers that have been through a traditional prepublication peer review process remain the most important means of communicating science today. Researchers depend on them to learn about latest advances in their fields and report own findings. The intentions are certainly noble: ensure methodological integrity comment potential significance experimental studies examination by panel objective, expert colleagues. In principle, this system enables move forward collective...
Increased mitogenic signaling by positive effectors such as Ras or Myc can trigger senescence in normal cells, a response believed to function tumor-suppressive mechanism. We report here the existence of checkpoint that monitors hypoproliferative imbalances. Normal human fibroblasts with one copy c-myc gene inactivated targeted homologous recombination switched an increased frequency telomere-independent senescent state mediated cyclin-dependent kinase inhibitor p16 INK4a . expression was...
Background Previous studies of major depressive disorder (MDD) have focused on abnormalities in the prefrontal cortex and medial temporal regions. There has been little investigation MDD midbrain subcortical regions central to reward/aversion function, such as ventral tegmental area/substantia nigra (VTA/SN), forebrain bundle (MFB). Methodology/Principal Findings We investigated microstructural integrity this circuitry using diffusion tensor imaging (DTI) 22 subjects compared them with...
Medicine, by its nature, is a multifaceted domain that requires the synthesis of information across various modalities. Medical generative vision-language models (VLMs) make first step in this direction and promise many exciting clinical applications. However, existing typically have to be fine-tuned on sizeable down-stream datasets, which poses significant limitation as medical applications data scarce, necessitating are capable learning from few examples real-time. Here we propose...
The leaves of the Coriandrum sativum plant, known as cilantro or coriander, are widely used in many cuisines around world. However, far from being a benign culinary herb, can be polarizing—many people love it while others claim that tastes smells foul, often like soap dirt. This soapy pungent aroma is largely attributed to several aldehydes present cilantro. Cilantro preference suspected have genetic component, yet date nothing about specific mechanisms. Here, we results genome-wide...
Abstract Since the start of COVID-19 pandemic, Asian Americans have been subjected to rising overt discrimination and violent hate crimes, highlighting health implications racism toward Americans. As are only group for whom cancer is leading cause death, these manifestations anti-Asian provoke question impact across continuum In this Commentary, we describe how myth “model minority” overlooks diversity Ignoring such in sociocultural trends, immigration patterns, socioeconomic status,...
Purpose Although many studies have shown a subjective preference for yellow lenses, there has been little success in determining the clinical nature of this benefit. Method Contrast sensitivity, color vision, accommodative-convergence, and visual acuity were measured group 20 young subjects along with rating their perception through clear control lenses (380-nm cut-off), (450-nm dark (511-nm orange (527-nm cut-off). Results A systematic detriment to vision was found occur increasing cut-off...
Large Language Models are typically trained with next-turn rewards, limiting their ability to optimize for long-term interaction. As a result, they often respond passively ambiguous or open-ended user requests, failing help users reach ultimate intents and leading inefficient conversations. To address these limitations, we introduce CollabLLM, novel general training framework that enhances multiturn human-LLM collaboration. Its key innovation is collaborative simulation estimates the...
Multi-agent AI systems powered by large language models (LLMs) are increasingly applied to solve complex tasks. However, these often rely on fragile, manually designed prompts and heuristics, making optimization difficult. A key challenge in optimizing multi-agent is acquiring suitable training data for specialized agents. We introduce SiriuS, a self-improving, reasoning-driven framework systems. Central our approach the construction of an experience library: repository high-quality...
Abstract Structural genomics efforts contribute new protein structures that often lack significant sequence and fold similarity to known proteins. Traditional structure-based methods may not be sufficient annotate the molecular functions of these structures. Techniques combine structural functional modeling can valuable for annotation. FEATURE is a flexible framework recognition sites in macromolecular Here, we present an overview main components framework, describe recent developments its...
Abstract Purpose: We determined the antiangiogenic and anticancer activity of VEGI-192, a new isoform TNFSF15 (VEGI, TL1), with Lewis lung cancer murine tumor model. Experimental Design: Recombinant human VEGI-192 was produced in Escherichia coli purified to apparent homogeneity. The protein given systemically via i.p., i.v., or s.c. injections tumor-bearing C57BL/6 mice. Tumor growth rates, animal survival general toxicity were determined. Effect on endothelial cell/smooth muscle cell ratio...
Alpha-1 antitrypsin deficiency (AATD) is an autosomal co-dominant condition that predisposes to emphysema, cirrhosis, panniculitis, and vasculitis. Underrecognition has prompted efforts enhance early detection testing of at-risk individuals. Direct-to-consumer (DTC) genetic represents additional method detection.
T cell quiescence is essential for maintaining a broad repertoire against large pool of diverse antigens from microbes and tumors, but the underlying molecular mechanisms remain largely unknown. We show here that CD8α critical maintenance CD8 + cells in physiologically quiescent state peripheral lymphoid organs. Upon inducible deletion CD8α, both naïve memory spontaneously acquired activation phenotypes subsequently died without exposure to specific antigens. PILRα was identified as ligand...
Despite its promises for biomedical applications, the lack of solubility in a physiological solution, limited molecular interactions with nucleic acids due to rigid backbone, and inefficient intracellular release limit use chitosan, natural cationic polysaccharide, gene delivery. In this study, flexible, aqueous-soluble aminoethoxy branch was conjugated primary hydroxyl group chitosan via an acid-cleavable ketal linkage, resulting acid-transforming (ATC) greatly increased aqueous solubility,...
While GPT-4V(ision) impressively models both visual and textual information simultaneously, it's hallucination behavior has not been systematically assessed. To bridge this gap, we introduce a new benchmark, namely, the Bias Interference Challenges in Visual Language Models (Bingo). This benchmark is designed to evaluate shed light on two common types of hallucinations language models: bias interference. Here, refers model's tendency hallucinate certain responses, possibly due imbalance its...
Deep neural networks often rely on spurious correlations to make predictions, which hinders generalization beyond training environments. For instance, models that associate cats with bed backgrounds can fail predict the existence of in other environments without beds. Mitigating is crucial building trustworthy models. However, existing works lack transparency offer insights into mitigation process. In this work, we propose an interpretable framework, Discover and Cure (DISC), tackle issue....
Microblogging platforms and other tools for videos, podcasts, virtual environments provide an untapped potential science conferences. Our experiment using FriendFeed to cover ISMB 2008 was educational surprisingly successful. We found that it enhanced our note-taking skills, allowed us compile notes from parallel sessions, attracted wider interest non-attendees, and, in addition the live aspect, generated a permanent archive of meeting. ISMB/ECCB 2009 will be held Stockholm. look forward new...
Abstract Structural genomics efforts have led to increasing numbers of novel, uncharacterized protein structures with low sequence identity known proteins, resulting in a growing need for structure-based function recognition tools. Our method, SeqFEATURE, robustly models functions described by motifs using structural representation. We built library that shows good performance compared other methods. In particular, SeqFEATURE demonstrates significant improvement over methods when and...
Answering real-world user queries, such as product search, often requires accurate retrieval of information from semi-structured knowledge bases or databases that involve blend unstructured (e.g., textual descriptions products) and structured entity relations information. However, previous works have mostly studied relational tasks separate topics. To address the gap, we develop STARK, a large-scale Semi-structure benchmark on Textual Relational Knowledge Bases. We design novel pipeline to...
Abstract Background The emergence of structural genomics presents significant challenges in the annotation biologically uncharacterized proteins. Unfortunately, our ability to analyze these proteins is restricted by limited catalog known molecular functions and their associated 3D motifs. Results In order identify novel motifs that may be with functions, we employ an unsupervised, two-phase clustering approach combines k-means hierarchical knowledge-informed cluster selection methods. We...
The Pacific Symposium on Biocomputing is an annual meeting whose topics are determined by proposals submitted members of the community. This document proposal for a session Open Science, consideration PSB in 2009.