Robust Instrumental Variables estimators to boost protein biomarkers discoveries using gene expression data

生物标志物可以定义为用于测量和评估疾病状态或对治疗反应的指标。例如，血浆低密度脂蛋白（LDL）胆固醇水平有助于监测患者的心血管健康状况。迄今为止，鉴定各种疾病的分子生物标志物（如蛋白质）的临床需求尚未得到满足，包括癌症、心力衰竭和慢性阻塞性肺疾病。这些指标有助于开发微创和及时有效的诊断工具，从而改善加拿大卫生系统的患者护理并降低成本。

Biomarkers can be defined as indicators used to measure and evaluate the state of a disease or a response to treatment. For example, plasma levels of low-density lipoprotein (LDL) cholesterol help to monitor patients' general cardiovascular health. To date, there is an unmet clinical need to identify molecular biomarkers (e.g., proteins) of various diseases, including cancer, heart failure, and chronic obstructive pulmonary disease. These indicators can yield to the development of minimally-invasive and time-effective diagnostic tools improving patients' care and decreasing costs in the Canadian health system. Novel technologies that can measure genes and proteins in the blood open new venues for these biomedical investigations. However, new statistical methods are needed to distil the information generated by these technologies. In the past, I have been working on the analysis of proteomics data aimed to develop clinical tests that can replace biopsies to monitor organ rejection in transplantation. My proposed research program offers appropriate computational approaches to find protein biomarkers, where classical statistical methods may fail to detect them, borrowing strength from related genomics expression. The new methodology will completely change the way that biomarkers are searched.