Support for New Bioinformatics Methods Development

支持新的生物信息学方法开发

• 批准号: 10281443
• 负责人: RUCHIR SHAH
• 金额: $ 21.03万
• 依托单位: SCIOME, LLC
• 依托单位国家: 美国
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-03-24 至 2021-03-23
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10281443
• 关键词: Bioinformatics; Cell Death; Cells; Chemical Exposure; Chemicals; Contractor; DNA Sequence; Development; Evaluation; Genes; Hepatocyte; Image Analysis; Measures; Methods; Output; Program Reviews; Programming Languages; Protocols documentation; Sampling; Series; Specific qualifier value; Stress; Toxic effect; base; bioinformatics tool; deep learning; differential expression; glyphosate; image processing; machine learning method; method development; programs; transcriptomics

New bioinformatics method development support includes, image analysis for glyphosate toxicity where deep-learning based image processing tmethods were used to discriminate between normal, stressed and cell-death conditions of HepaRG cells and primary hepatocytes; Evidence tagging protocols were develop for evidence mapping for the OHAT group; an evaluation of existing tagging methods was performed currently available in the SWIFT-Review program; Machine Learning methods were used for Document tagging activity exploring alternative to the keyword-based tagging strategy currently used in SWIFT-Review.

新的生物信息学方法开发支持包括草甘膦毒性的图像分析，其中基于深度学习的图像处理方法用于区分HepaRG细胞和原代肝细胞的正常、应激和细胞死亡条件;为OHT组的证据映射开发证据标记协议;对SWIFT-Review程序中目前可用的现有标记方法进行评估;在文件标记活动中使用了机器学习方法，以探索SWIFT-Review目前使用的基于关键字的标记策略的替代方案。