Biology subjects

Terry, J.

Publications and source records attributed to Terry, J..

5 recordsLinked to original sources

bioRxiv2026

MOSAIC: Learning Graph Node Embeddings from Spectrally Isolated Dominant Subspaces of Accumulated Diffusion Operators

Graph embedding transforms complex networks into low-dimensional representations for analysis and downstream machine learning. Existing methods often rely on eigenvalue ordering, random-walk sampling, or neural optimization. We introduce MOSAIC, Multistep Operator Spectral Accumulation with Isolated Components, an unsupervised framework that constructs an accumulated diffusion operator from successive powers of a graph transition matrix. MOSAIC uses Krylov subspace analysis to identify dominant spectral components according to their convergence behavior. This approach emphasizes persistent and spectrally separated diffusion modes rather than selecting embedding coordinates solely by eigenvalue magnitude. We evaluated MOSAIC using a biomedical graph projection derived from the Data Distillery Knowledge Graph that contained 5,171 human genes and 10,147 phenotypes. Across embedding dimensions from 16 to 128, MOSAIC achieved node-type classification AUROCs from 0.9863 to 0.9926 and exceeded FastRP, Node2Vec, HashGNN, and GraphSAGE. For gene-phenotype link prediction, validation and test associations were withheld before embedding generation. Under progressive removal of all edges or specific edge classes, MOSAIC retained more node-label information than FastRP, including after extreme graph sparsification. In a focused DICER1 analysis, 106 of 137 documented phenotype associations received scores above 0.90, demonstrating concordance with known pleiotropic biology and prioritizing additional relationships for independent evaluation. These results support MOSAIC as a fully unsupervised framework for learning informative graph representations. Although demonstrated using a biomedical graph projection, MOSAIC is domain independent and can be applied to other complex networks

bioinformatics↗

bioRxiv2025

The Data Distillery: A Graph Framework for Semantic Integration and Querying of Biomedical Data

The Data Distillery Knowledge Graph (DDKG) is a framework for semantic integration and querying of biomedical data across domains. Built for the NIH Common Fund Data Ecosystem, it supports translational research by linking clinical and experimental datasets in a unified graph model. Clinical standards such as ICD-10, SNOMED, and DrugBank are integrated through UMLS, while genomics and basic science data are structured using ontologies and standards such as HPO, GENCODE, Ensembl, STRING, and ClinVar. The DDKG uses a property graph architecture based on the UBKG infrastructure and supports ontology-based ingestion, identifier normalization, and graph-native querying. The system is modular and can be extended with new datasets or schema modules. We demonstrate its utility for informatics queries across eight use cases, including regulatory variant analysis, tissue-specific expression, biomarker discovery, and cross-species variant prioritization. The DDKG is accessible via a public interface, a programmatic API, and downloadable builds for local use.

bioinformatics↗

bioRxiv2024

Analysis of a hypomorphic mei-P26 mutation reveals developmental control of CO patterning mechanisms

Female gametogenesis in Drosophila requires differentiation and mitotic division of germ cells, acquisition of oocyte fate, and entry into meiosis. Each of these processes is well understood individually; however, little is known about the mechanisms that ensure proper temporal integration of germ cell differentiation and meiotic chromosome dynamics. Here, we take advantage of a hypomorphic mutation in mei-P26, a well-characterized gene with multiple diverse functions in germ cell development, to determine the consequences of disrupting the coordination between development and meiosis. While null mutations in mei-P26 lead to tumorous ovaries, the hypomorphic allele mei-P261 allows sufficient germ cell differentiation and fertility to support analysis of meiotic chromosome dynamics. Unlike wildtype germaria, 60% of cysts in mei-P261 germaria co-express the differentiation factor Bag of marbles (Bam) and the oocyte specification factor Orb, suggesting that mitotic division is delayed. In this context, the synaptonemal complex rarely assembles into full length continuous tracks and instead is missing or present only as foci. Despite these phenotypes, meiotic double-strand breaks still form and are repaired as crossovers, but the crossovers are mis-patterned and form in centromere proximal regions rather than chromosome arms. The strength of crossover interference is significantly reduced and the centromere effect is lost, but crossover assurance is intact and the meiosis-specific machinery is used to form crossovers. We suggest a model where the failure to exit mitosis in a timely fashion causes cells to enter meiosis while still receiving mitotic signals, resulting in abnormal meiotic chromosome dynamics and impaired crossover formation. Article summaryFemale gametogenesis in Drosophila requires the precise temporal integration of germ cell differentiation and meiotic entry. Using a hypomorphic mutation in mei-P26, we investigated how disrupting this coordination affects meiotic chromosome dynamics. In mei-P261 mutants, delayed mitotic exit causes cells to enter meiosis while still receiving mitotic signals. This results in fragmented synaptonemal complex and the loss of some but not all CO patterning mechanisms. These results suggest that timely mitotic exit is critical for establishing the proper regulatory landscape required for meiotic recombination.

genetics↗

bioRxiv2024

A novel dataset of 2,362 equine fecal microbiomes from eight veterinary teaching hospital on three continents reveals dominant effects of geography, breed, and disease

Horses and other equids are reliant on the gut microbiome for health, and studies have reported associations between certain clinical conditions and features of the fecal microbiome. However, research to date on the equine fecal microbiome has often relied on small sample sizes collected from single and relatively localized geographic regions. Previous work largely employs single timepoint analyses, or horses selected based on limited health criteria. To address these issues and expand our understanding of the core microbiome in health, and the changes associated with adverse outcomes, the Equine Gut Group (EGG) has collected and performed 16S rRNA sequencing on 2,362 fecal samples from 1,190 healthy and affected horses. Here we present the EGG database and demonstrate its utility in characterizing the equine microbiome in health and acute gastrointestinal disease. The EGG 16S rRNA database is a valuable resource to study the equine microbiome and its role in equine health.

microbiology↗

bioRxiv2022

Assessment of the Human Placental Microbiome in Early Pregnancy

Bacteria derived from the maternal circulation have been suggested to seed the human placenta during development leading to an intrinsic placental microbiome. This concept has become controversial as numerous studies suggest that the apparent placental microbiome is mostly, if not completely, comprised of contaminants. If the maternal circulation seeds the placenta then there should be an increase in abundance and diversity of detectable bacteria with onset of maternal perfusion of the placenta around 10 weeks gestational age; however, if only contaminants are present then there should be no significant evolution of the placental microbiome with increasing gestational age. This pilot study addresses whether bacterial abundance and diversity increases in human placenta and whether there is an associated shift in the immunophenotype of the decidual immune cell complement before and after initiation of placental perfusion. Human placental and decidual tissue from 5-19 weeks gestational age are assessed by quantitative 16S polymerase chain reaction (PCR), 16S gene sequencing, and immunological flow cytometry studies. A weak positive correlation between placental bacterial abundance and gestational age is identified but is not statistically significant. No significant changes in bacterial diversity are found with increasing gestational age. The proportion of decidual activated memory T helper cells increases with gestational age but no change was observed in other lymphocyte subsets. This pilot study does not strongly support bacterial colonization of the placenta after initiation of maternal perfusion; however, the minor trends towards increases in bacterial abundance and activated memory T helper cells may represent an early stage of this process. Additional investigations in larger cohorts are warranted.

physiology↗