Towards an autonomous in silico researcher: Using logic modelling to automate the explanation of unexpected results

走向自主的计算机研究人员：使用逻辑建模自动解释意外结果

• 批准号: 2241937
• 金额: --
• 依托单位: Queen Mary University of London
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2241937
• 关键词: Towards; autonomous; silico; researcher; Using

Biological experiments often produce unexpected results, i.e. results that contradict the researcher's existing understanding. Such results can be revolutionary, revealing new knowledge or leading to new testable hypotheses, but they can also be due to gaps in the researcher's prior knowledge, or experimental error. Explaining unexpected results is therefore critically important. However, today's massive datasets contain so many unexplained results only a minority can be investigated manually, potentially leaving new discoveries unnoticed. This project aims to develop methods that combine experimental data, metadata and prior knowledge to automatically explain all unexpected datapoints within a large dataset, hastening biological discoveries.

生物学实验经常产生意想不到的结果，即与研究人员现有的理解相矛盾的结果。这样的结果可能是革命性的，揭示了新的知识或导致新的可检验的假设，但它们也可能是由于研究人员的先验知识或实验误差的差距。因此，解释意外结果至关重要。然而，今天的大规模数据集包含如此多的无法解释的结果，只有少数可以手动调查，可能会忽略新的发现。该项目旨在开发结合联合收割机实验数据、元数据和先验知识的方法，以自动解释大型数据集中所有意想不到的数据点，加速生物学发现。