Kiley Graim
A5009087038- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Occupational and environmental lung diseases
- Medical Imaging and Pathology Studies
- Genetics, Bioinformatics, and Biomedical Research
- Cancer Genomics and Diagnostics
- Ferroptosis and cancer prognosis
- Radiomics and Machine Learning in Medical Imaging
- Sepsis Diagnosis and Treatment
- Prostate Cancer Treatment and Research
- Molecular Biology Techniques and Applications
- Veterinary Oncology Research
- Cancer Research and Treatments
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- Metabolomics and Mass Spectrometry Studies
- Cancer, Lipids, and Metabolism
- MicroRNA in disease regulation
- Machine Learning in Healthcare
- Gene Regulatory Network Analysis
- Genomics and Phylogenetic Studies
- Biomedical Text Mining and Ontologies
- Prostate Cancer Diagnosis and Treatment
- Cancer, Hypoxia, and Metabolism
- Cancer Cells and Metastasis
University of Florida
2022-2025
University of Florida Health
2024-2025
Florida College
2025
UF Health Cancer Center
2024
Simons Foundation
2018-2020
Princeton University
2018-2020
University of California, Santa Cruz
2013-2018
Santa Cruz County Office of Education
2014
Colorado State University
2013
Automated annotation of protein function is challenging. As the number sequenced genomes rapidly grows, overwhelming majority products can only be annotated computationally. If computational predictions are to relied upon, it crucial that accuracy these methods high. Here we report results from first large-scale community-based critical assessment (CAFA) experiment. Fifty-four representing state art for prediction were evaluated on a target set 866 proteins 11 organisms. Two findings stand...
Abstract Malignant pleural mesothelioma (MPM) is a highly lethal cancer of the lining chest cavity. To expand our understanding MPM, we conducted comprehensive integrated genomic study, including most detailed analysis BAP1 alterations to date. We identified histology-independent molecular prognostic subsets, and defined novel subtype with TP53 SETDB1 mutations extensive loss heterozygosity. also report strong expression immune-checkpoint gene VISTA in epithelioid strikingly higher than...
It remains unclear whether causal, rather than merely correlational, relationships in molecular networks can be inferred complex biological settings. Here we describe the HPN-DREAM network inference challenge, which focused on learning causal influences signaling networks. We used phosphoprotein data from cancer cell lines as well silico a nonlinear dynamical model. Using data, scored more 2,000 submitted by challenge participants. The spanned 32 contexts and were terms of validity with...
Significance Aggressive cancers often possess functional and molecular traits characteristic of normal stem cells. It is unclear if aggressive phenotypes prostate cancer molecularly resemble cells residing within the human prostate. Here, we transcriptionally profiled epithelial populations from show that enriched for a basal cell signature. Within metastases, histological subtypes had varying enrichment signature, with small neuroendocrine carcinoma being most cell-like. We further found...
Abstract The GPT-4 large language model (LLM) and ChatGPT chatbot have emerged as accessible capable tools for generating English-language text in a variety of formats. has previously performed well when applied to questions from multiple standardized examinations. However, further evaluation trustworthiness accuracy responses across various knowledge domains is essential before its use reference resource. Here, we assess performance on nine graduate-level examinations the biomedical...
Recent genomic analyses have provided substantial evidence for past periods of gene flow from polar bears (Ursus maritimus) into Alaskan brown arctos), with some suggesting a link between climate change and introgression. However, because it has mainly been possible to sample the present day, timing, frequency, evolutionary significance this admixture remains unknown. Here, we analyze DNA three additional geographically distinct bear populations, including two that lived temporally close...
Vast amounts of molecular data are being collected on tumor samples, which provide unique opportunities for discovering trends within and between cancer subtypes. Such cross-cancer analyses require computational methods that enable intuitive interactive browsing thousands samples based their similarity. We created a portal called TumorMap to assist in exploration statistical interrogation high-dimensional complex "omics" an easily interpretable way. In the TumorMap, arranged hexagonal grid...
Combining heterogeneous sources of data is essential for accurate prediction protein function. The task complicated by the fact that while sequence-based features can be readily compared across species, most other are species-specific. In this paper, we present a multi-view extension to GOstruct, structured-output framework function annotation proteins. extended learn from disparate sources, with each source provided in form kernel. Our empirical results demonstrate able utilize all...
The purpose of this study was to investigate changes in the lipidome patients with sepsis identify signaling lipids associated poor outcomes that could be linked future therapies. Adult were enrolled within 24h recognition. Patients meeting Sepsis-3 criteria from emergency department or intensive care unit and blood samples obtained. Clinical data collected rapid recovery, chronic critical illness (CCI), early death adjudicated by clinicians. Lipidomic analysis performed on two platforms,...
Abstract OBJECTIVE To compare pedigree documentation and genetic test results to evaluate whether user-provided photographs influence the breed ancestry predictions of direct-to-consumer (DTC) tests for dogs. ANIMALS 12 registered purebred pet dogs representing different breeds. METHODS Each dog owner submitted 6 buccal swabs, 1 each DTC testing companies. Experimenters sample per manufacturer instructions. For half dogs, registration included a photograph DNA donor. other were swapped...
Gold standard genomic datasets severely under-represent non-European populations, leading to inequities and a limited understanding of human disease. Therapeutics outcomes remain hidden because we lack insights that could be gained from analyzing ancestrally diverse data. To address this significant gap, present PhyloFrame, machine learning method for equitable precision medicine. PhyloFrame corrects ancestral bias by integrating functional interaction networks population genomics data with...
Abstract The emerging field of comparative oncology may provide a novel solution to rare human cancer studies, as genomic resources, reference genomes, and non-human model sequencing data are proliferating rapidly. Studying across diverse array species provides unique opportunity interrogate factors underpinning cancer, facilitating disease modeling in the setting spontaneous tumors complicated by comorbidities metastases. Unfortunately, spontaneously developed mammalian models have been...
Abstract Rapid advancements in genomic sequencing technologies have revolutionized the study and treatment of human diseases. Even with wealth information available, successful application developed treatments is often limited to populations represented original study, suggesting that these studies are overfitting their data thus limiting scientific understanding complex traits Human population-specific variants response evolutionary changes as a result migrations other stressors which can...
Abstract Background Cholesterol metabolism is dysregulated in sepsis contributing to patient heterogeneity. Subphenotypes displaying lower lipoprotein levels and higher mortality (HYPO) or (NORMO), were described. We developed a simplified clinical algorithm for bedside subphenotype recognition. Methods analyzed data from four prospective studies (internal dataset), focusing on HYPO NORMO subphenotypes. A 1,000-tree Random Forest classifier logistic regression models built, using features...
Leiomyosarcoma (LMS) is a rare, aggressive, mesenchymal tumor. Subsets of LMS have been identified to harbor genomic alterations associated with homologous recombination deficiency (HRD); particularly in BRCA2. Whereas loss heterozygosity (gLOH) has used as surrogate marker HRD other solid tumors, the prognostic or clinical value gLOH (gLOH-LMS) remains poorly defined. We explore drivers gLOH-LMS and their import. Although distribution scores are similar that carcinomas, outside BRCA2, there...
This is a study of lipid metabolic gene expression patterns to discover precision medicine for sepsis.
Understanding the changes in diverse molecular pathways underlying development of breast tumors is critical for improving diagnosis, treatment, and drug development. Here, we used RNA-profiling canine mammary (CMTs) coupled with a robust analysis framework to model human cancer. Our study leveraged key advantage model, frequent presence multiple naturally occurring at thus providing samples spanning normal tissue benign malignant from each patient. We showed cancer signals, both expression...
We introduce a novel method called Prophetic Granger Causality (PGC) for inferring gene regulatory networks (GRNs) from protein-level time series data. The uses an L1-penalized regression adaptation of to model protein levels as function time, stimuli, and other perturbations. When combined with data-independent network prior, the framework outperformed all methods submitted HPN-DREAM 8 breast cancer inference challenge. Our investigations reveal that PGC provides complementary information...
Patient stratification to identify subtypes with different disease manifestations, severity, and expected survival time is a critical task in cancer diagnosis treatment. While approaches using various biomarkers (including high-throughput gene expression measurements) for patient-to-patient comparisons have been successful elucidating previously unseen subtypes, there remains an untapped potential of incorporating genotypic phenotypic data discover novel or improved groupings. Here, we...
Abstract Circulating extracellular vesicles (EVs) have gained significant attention for discovering tumor biomarkers. However, isolating EVs with well-defined homogeneous populations from complex biological samples is challenging. Different isolation methods been found to derive different EV carrying molecular contents, which confounds current investigations and hinders subsequent clinical translation. Therefore, standardizing building a rigorous assessment of isolated quality associated...