Resolving Methodological Challenges in Genomics Research: Causality, Risk Prediction, and Reproducibility

项目摘要 随着高通量测序技术的问世，科学界现在 能够在生物和分子水平上研究复杂的表型。不过， 进行对照试验和随机试验仍然相当困难 在不同水平上调查表型之间的因果关系。因此，它是至关重要的 重要的是根据观测数据进行因果推断。在这个项目中，我们将 发展计算方法以促进对因果分子的系统研究 复杂疾病过程的潜在机制。具体地说，我们将针对三个杰出的 科学问题：i)复杂疾病分子机制的随意推断；ii)分析 利用基因组信息和因果分子机制进行风险预测的方法， 以及iii)高通量基因组实验的重复性的统计评估。最后， 我们将制作方便用户使用的计算软件包，并向广大用户提供 生物和医学科学家的社区。

Project Summary With the availability of the high-through sequencing technology, the scientific community is now able to investigate complex phenotypes at both organismal and molecular levels. Nevertheless, it is still considerably difficult to perform controlled experiments and randomized trials to investigate the causal relationships between phenotypes at different levels. It is therefore critically important to perform causal inference based on the observational data. In this project, we will develop computational methods to facilitate systematic investigation of causal molecular mechanisms underlying complex disease process. Specifically, we will target three outstanding scientific issues: i) casual inference of molecular mechanisms of complex diseases; ii) analytic approaches for risk prediction utilizing genomic information and causal molecular mechanisms, and iii) statistical assessment of reproducibility in high-throughput genomic experiments. Finally, we will build user-friendly computational software packages and make them available to the broad community of biological and medical scientists.