The correlation of phenotypic outcomes with genetic variation and environmental factors is a core pursuit in biology and biomedicine. Numerous challenges impede our progress: patient phenotypes may not match known diseases, candidate variants may be in genes that have not been characterized, model organisms may not recapitulate human or veterinary diseases, filling evolutionary gaps is difficult, and many resources must be queried to find potentially significant genotype-phenotype associations. Non-human organisms have proven instrumental in revealing biological mechanisms. Advanced informatics tools can identify phenotypically relevant disease models in research and diagnostic contexts. Large-scale integration of model organism and clinical research data can provide a breadth of knowledge not available from individual sources and can provide contextualization of data back to these sources. The Monarch Initiative (monarchinitiative.org) is a collaborative, open science effort that aims to semantically integrate genotype-phenotype data from many species and sources in order to support precision medicine, disease modeling, and mechanistic exploration. Our integrated knowledge graph, analytic tools, and web services enable diverse users to explore relationships between phenotypes and genotypes across species.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5210586 | PMC |
| http://dx.doi.org/10.1093/nar/gkw1128 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Article Synopsis
- - Large language models (LLMs) are being tested for their ability to help diagnose genetic diseases, but their evaluation is complicated due to how they generate unstructured responses.
- - Researchers benchmarked LLMs against 5,213 case reports using established phenotypic criteria and compared their performance to a traditional diagnostic tool, Exomiser.
- - The best-performing LLM correctly diagnosed cases 23.6% of the time, while Exomiser achieved 35.5%, indicating that while LLMs are improving, they still lag behind conventional bioinformatics methods and need further research for effective integration into diagnostic processes.
FastHPOCR: pragmatic, fast, and accurate concept recognition using the human phenotype ontology.
Bioinformatics
July 2024
Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Charitéplatz 1, 10117 Berlin, Germany.
Motivation: Human Phenotype Ontology (HPO)-based phenotype concept recognition (CR) underpins a faster and more effective mechanism to create patient phenotype profiles or to document novel phenotype-centred knowledge statements. While the increasing adoption of large language models (LLMs) for natural language understanding has led to several LLM-based solutions, we argue that their intrinsic resource-intensive nature is not suitable for realistic management of the phenotype CR lifecycle. Consequently, we propose to go back to the basics and adopt a dictionary-based approach that enables both an immediate refresh of the ontological concepts as well as efficient re-analysis of past data.
View Article and Find Full Text PDFNode-degree aware edge sampling mitigates inflated classification performance in biomedical random walk-based graph representation learning.
Bioinform Adv
March 2024
The Jackson Laboratory for Genomic Medicine, CT 06032, United States.
Motivation: Graph representation learning is a family of related approaches that learn low-dimensional vector representations of nodes and other graph elements called embeddings. Embeddings approximate characteristics of the graph and can be used for a variety of machine-learning tasks such as novel edge prediction. For many biomedical applications, partial knowledge exists about positive edges that represent relationships between pairs of entities, but little to no knowledge is available about negative edges that represent the explicit lack of a relationship between two nodes.
View Article and Find Full Text PDFStructured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES): a method for populating knowledge bases using zero-shot learning.
Bioinformatics
March 2024
Biosystems Data Science, Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA 94720, United States.
Motivation: Creating knowledge bases and ontologies is a time consuming task that relies on manual curation. AI/NLP approaches can assist expert curators in populating these knowledge bases, but current approaches rely on extensive training data, and are not able to populate arbitrarily complex nested knowledge schemas.
Results: Here we present Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES), a Knowledge Extraction approach that relies on the ability of Large Language Models (LLMs) to perform zero-shot learning and general-purpose query answering from flexible prompts and return information conforming to a specified schema.
Migratory animals follow seasonal cycles comprising linked phases often with different habitat requirements and demographic processes. Conservation of migratory species therefore must consider the full seasonal cycle to identify points limiting population viability. For western monarch butterflies, which have experienced significant declines, early spring is considered a critical period in the annual population cycle.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!