Statistical Inference in Biology: Intuition from a Historical Perspective

Undergraduate / Graduate

Biology is undergoing a substantial transition. With the advent of new technologies that enable the collection of massive amounts of data coupled with advances in computational frameworks and hardware, Biology is squarely in the era of statistical learning-inferring models of how Biology works from statistics rather than a deep understanding of complex mechanism. How did this shift come about? Why did this shift come about? How can we really understand how to deal with 'complexity' in Biology? This course is meant to provide students with an intuition regarding statistical inference as an emerging philosophy for studying Biology. As such, this course is centered around key papers that thematically frame (i) each week and (ii) several consecutive lectures of the course. We begin by introducing the concept of reductionism in Biology and its limitations. We move on to describing complexity and how the scientific need for statistical inference frameworks follows from this description. We discuss how these statistical inference frameworks have been implemented and how results from these frameworks have shaped our understanding of complex biological systems through extensive validation efforts and cutting-edge design strategies geared toward engineering synthetic biological systems. Finally, we discuss what key mathematical aspects define statistical inference, including a very short discussion regarding machine-learning and artificial intelligence applications in Biology.