岗位介绍
您将加入克拉克博士的实验室，该实验室隶属于德国遗传学与遗传学研究中心 (DKFZ) 计算基因组学和系统遗传学部门 (https://steglelab.org)。该项目与慕尼黑亥姆霍兹中心 TARGET-AI 项目的主要合作伙伴以及加州大学伯克利分校和德国人类基因组-表型组档案馆的支持团队合作开展。该职位还将与当地充满活力的数据科学和机器学习生态系统紧密相连。
你的任务
Brian Clarke 博士的研究小组正在寻找一位积极进取的博士生加入一个雄心勃勃的项目，该项目旨在构建机器学习和深度学习模型，用于研究人类疾病的遗传学。TARGET-AI 项目由亥姆霍兹人工智能项目资助，将汇集多个研究小组的专业知识，推动深度学习/人工智能领域最先进的技术与严谨的统计分析相结合，而我们团队在该领域拥有丰富的成功经验（请参阅下文近期发表的成果）。
TARGET-AI 项目旨在应用深度学习和贝叶斯建模的前沿技术，深入了解疾病的遗传基础，并改进遗传风险预测。我们力求以团队先前积累的专业知识和方法（见下文）为基础，包括整合贝叶斯神经网络和生物序列模型（包括大型 DNA 和蛋白质语言模型）的架构和原理。该项目还旨在开发一个联邦学习框架原型，以便对多个大型生物样本库数据集进行荟萃分析，旨在提高我们对疾病复杂遗传病因的敏感性，并由此发现新的治疗靶点。
最近的相关出版物：
Clarke, B.，Holtkamp, E. 等。利用深度集网络整合变异注释，提升罕见变异关联检测能力。《自然遗传学》56, 2271–2280 (2024)。https ://doi.org/10.1038/s41588-024-01919-z
Nappi, A. 等人。多重注释的贝叶斯聚合增强了罕见变异关联检测。bioRxiv (2025)。https ://doi.org/10.1101/2025.03.02.641062
You will be affiliated with the laboratory of Dr. Clarke, which is part of the Division of Computational Genomics and Systems Genetics at DKFZ (https://steglelab.org). The project is executed in collaboration with the main partner of the TARGET-AI project at Helmholtz Munich, as well as supporting groups at the University of California, Berkeley and the German Human Genome-Phenome Archive. The position will also be connected to a vibrant local ecosystem for data science and machine learning.
YOUR TASKS
The research group of Dr. Brian Clarke is looking for a highly motivated doctoral student to join an ambitious project aimed at building machine and deep learning models to study the genetics of human disease. Funded as part of the Helmholtz AI program, the project TARGET-AI will bring together expertise from multiple research groups to advance the state-of-the-art in combining the most advanced techniques from deep learning/AI with rigorous statistical analyses, an area in which our group has a track record of success (see recent publications below).
The TARGET-AI project seeks to apply leading-edge techniques from deep learning and Bayesian modeling to gain insights into the genetic underpinnings of disease and improve genetic risk prediction. We seek to build on previous expertise and methods devised by our teams (see below), including incorporating architectures and principles from Bayesian neural networks and biological sequence models, including large DNA and protein language models. The project also aims to develop a prototype federated learning framework to enable meta-analysis of multiple large biobank datasets, all of this with the aim of increasing our sensitivity to unravel the complex genetic causes of disease and, in so doing, identify new therapeutic targets.
Recent relevant publications:
Clarke, B., Holtkamp, E. et al. Integration of variant annotations using deep set networks boosts rare variant association testing. Nature Genetics 56, 2271–2280 (2024). https://doi.org/10.1038/s41588-024-01919-z
Nappi, A. et al. Bayesian Aggregation of Multiple Annotations Enhances Rare Variant Association Testing. bioRxiv (2025). https://doi.org/10.1101/2025.03.02.641062
岗位要求
成功的申请人应拥有计算机科学、统计学、数学、物理学和/或工程学硕士学位或同等学历，或拥有生物科学学位并具有定量分析经验。
我们正在寻找涵盖机器学习、数学或统计学等领域的人才和专业知识。欢迎具备项目重点领域（包括深度学习、贝叶斯建模和联邦学习）相关经验的申请者。
要求 具备 良好 的 英语 口头 和 书面 沟通 能力 .
我们的团队重视创造力、严谨的科学思维、友善的性格以及积极的团队精神。我们欢迎拥有不同专业和个人背景的候选人申请。
The successful applicant should hold a master's degree or equivalent qualification in computer science, statistics, mathematics, physics, and/or engineering, or a degree in biological science with demonstrated experience in quantitative analysis.
We are looking for a range of talents and expertise, including in machine learning, mathematics, or statistics. Experience with one of the focus areas of the project, including deep learning, Bayesian modeling, and federated learning, is welcome.
Effective oral and written communication skills in English are required .
Our group values creativity, a rigorous scientific mindset, kindness, and positive team spirit. We encourage candidates from diverse professional and personal backgrounds to apply.
咨询
Dr Brian Clarke
+49 (0)6221/42-4725