Applying machine learning models to genome data to understand the evolution of drug resistance from virus to cancer evolution

将机器学习模型应用于基因组数据，以了解从病毒到癌症的耐药性演变

• 批准号: 2453134
• 金额: --
• 依托单位: University of Glasgow
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2453134
• 关键词: Applying; machine; learning; models; genome

Studentship strategic priority area: Mathematics, statistics and computationKeywords: Virus evolution, cancer evolution, machine learning, genomicsTreatment of cancer and chronic infectious diseases often fail due to the evolution of resistance to therapy. Underpinning this phenomena is the generation of changes (mutations) in their genetic material. Mutations generate high levels of differences in the genomes of cancer cells or intra-patient virus populations that leads to their ability to evolve in response to drugs. Recent advances in genome sequencing have revealed genomic alterations that drive cancer progression and pathogen infection. These data give insight into the diseases' underlying evolutionary dynamics which follow predictions of both Darwin's theory of evolution and Motoo Kimura's theory of molecular evolution. Yet, how evolutionary dynamics interact with mutational processes, and whether these processes can predict clinical outcome is largely unknown. Due to the variety and complexity of genomic alterations observed across human, cancer and virus evolution, unified mathematical equations of evolution are often intractable. We propose to leverage state of the art machine learning methods applied to large scale genome sequencing data sets to build biologically informed data-driven models of evolutionary dynamics. These models permit efficient data analysis that account for the variety and complexity of genomic alterations observed across human, cancer and virus evolution. They will infer the life histories of disease processes and predict disease progression and effects of interventions. Early prediction of resistance to therapies is essential to maximising the potency of interventions and switching treatments when necessary. The student will be trained in a combination of data science and bioinformatics, with substantial elements of computation, programming and statistics/machine learning.

学生战略优先领域：数学、统计和计算关键词：病毒进化、癌症进化、机器学习、基因组学癌症和慢性传染病的治疗常常因治疗耐药性的进化而失败。这种现象的基础是其遗传物质发生变化（突变）。突变会在癌细胞或患者体内病毒群体的基因组中产生高度差异，从而导致它们能够响应药物而进化。基因组测序的最新进展揭示了驱动癌症进展和病原体感染的基因组改变。这些数据让我们深入了解疾病的潜在进化动力学，这些进化动力学遵循达尔文进化论和木村元夫分子进化论的预测。然而，进化动力学如何与突变过程相互作用，以及这些过程是否可以预测临床结果在很大程度上尚不清楚。由于在人类、癌症和病毒进化过程中观察到的基因组改变的多样性和复杂性，统一的进化数学方程通常很棘手。我们建议利用应用于大规模基因组测序数据集的最先进的机器学习方法来构建生物信息数据驱动的进化动力学模型。这些模型可以进行有效的数据分析，解释在人类、癌症和病毒进化过程中观察到的基因组改变的多样性和复杂性。他们将推断疾病过程的生命史并预测疾病进展和干预措施的效果。早期预测治疗耐药性对于最大限度地发挥干预措施的效力并在必要时更换治疗至关重要。学生将接受数据科学和生物信息学相结合的培训，其中包括计算、编程和统计/机器学习的实质内容。