Liam Beckman

ORCID: 0009-0008-4585-9181
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Bioinformatics and Genomic Networks
  • Distributed and Parallel Computing Systems
  • Cloud Computing and Resource Management
  • Health, Environment, Cognitive Aging
  • Cardiovascular Health and Risk Factors
  • Advanced Proteomics Techniques and Applications
  • AI in cancer detection
  • Computational Drug Discovery Methods
  • Gene expression and cancer classification

Oregon Health & Science University
2021-2024

University of Portland
2024

The Reactome Knowledgebase (https://reactome.org), an Elixir core resource, provides manually curated molecular details across a broad range of physiological and pathological biological processes in humans, including both hereditary acquired disease processes. are annotated as ordered network transformations single consistent data model. thus functions digital archive human tool for discovering functional relationships such gene expression profiles or somatic mutation catalogs from tumor...

10.1093/nar/gkab1028 article EN cc-by Nucleic Acids Research 2021-10-14

Abstract Cancer early detection is one of the most critical areas cancer research, as it offers greatest potential for improving patient outcomes. The International Alliance Early Detection (ACED) a global partnership world-leading research institutions in UK and US, established 2019 to accelerate revolutionize this field. ACED brings together expertise Canary Center at Stanford University, University Cambridge, Knight Institute Oregon Health Sciences College London, Manchester, with...

10.1158/1538-7445.am2024-3558 article EN Cancer Research 2024-03-22

The Global Alliance for Genomics and Health (GA4GH) Task Execution Service (TES) API is a standardized schema describing executing batch execution tasks. It provides common way to submit manage tasks variety of compute environments, including on premise High Performance Compute Throughput Computing (HPC/HTC) systems, Cloud computing platforms, hybrid environments. TES designed be flexible extensible, allowing it adapted wide range use cases, such as "bringing the data" solutions federated...

10.48550/arxiv.2405.00013 preprint EN arXiv (Cornell University) 2024-02-08

The accuracy of machine learning methods is often limited by the amount training data that available. We proposed to improve regimes augmenting datasets with synthetically generated samples. present a method for synthesizing gene expression samples and test system's capabilities improving categorical prediction cancer subtypes. developed SyntheVAEiser, variational autoencoder based tool was trained tested on over 8000 have shown this technique can be used augment tasks increase performance...

10.1186/s13059-024-03431-3 article EN cc-by-nc-nd Genome biology 2024-12-18
Coming Soon ...