Machine learning for coping with uncertain biological data

机器学习和生物信息学方法可用于发现有关复杂生物过程（例如衰老生物学）的数据中的新模式。然而，生物学中的数据经常涉及不确定性，这种不确定性往往导致不太可靠的预测模型。在生物学中，这种不确定性的例子是基因精确功能的不确定性，或者蛋白质是否与另一蛋白质相互作用的不确定性。这个跨学科的博士项目将专注于开发新的监督机器学习方法，以应对生物数据中的不确定性，特别是关于衰老相关基因和蛋白质的特性和功能的数据。该项目的总体目标是开发新的机器学习方法，不仅可以提高预测性能，还可以发现有关衰老生物学的新模式或新见解。

Machine learning and bioinformatics methods can be used to discover new patterns in data about complex biological processes (e.g. the biology of ageing). However, data in biology often involves uncertainty, and this uncertainty often leads to less reliable predictive models. Examples of such uncertainty in biology are the uncertainty about the precise functions of a gene or the uncertainty about whether or not a protein interacts with another protein. This inter-disciplinary PhD project will focus on developing new supervised machine learning methods for coping with uncertainty in biological data, particularly data about the properties and functions of ageing-related genes and proteins. The overall aim of the project is to develop new machine learning methods that not only improve predictive performance, but also lead to the discovery of new patterns or new insights about the biology of ageing.