Philosophy of Biology - The Scientific Discovery Process

The Scientific Discovery Process

Research in biology continues to be less guided by theory than it is in other sciences. This is especially the case in the context of life sciences, where the availability of high throughput screening techniques for the different omics fields and the perceived complexity, makes the science predominantly data driven. This data-intensive scientific discovery is by some considered to be the fourth paradigm, after empiricism, theory and computer simulation. Others reject the idea that data driven research is about to replace theory. As Krakauer et al. put it: "machine learning is a powerful means of preprocessing data in preparation for mechanistic theory building, but should not be considered the final goal of a scientific inquiry." In regard to cancer biology, Raspe et al. state: "A better understanding of tumor biology is fundamental for extracting the relevant information from any high throughput data."

Theory in biology is less strict formalized as it is in physics. Besides 1) the classic physics way of mathematical-analytical, there is 2) statistical based, 3) computer simulation and 4) conceptual/verbal theorizing/modeling. Dougherty and Bittner state that in order for biology to progress as a science, it has to move to more rigorous mathematical modeling, or otherwise risk to be "empty talk".

In tumor biology research, the characterization of cellular signaling processes has largely focused on identifying the function of individual genes and proteins. Janes showed however the context-dependent nature of signaling driving cell decisions demonstrating the need for a more system based approach . The lack of attention for context dependency in preclinical research is also illustrated by the observation that preclinical testing rarely includes predictive biomarkers that, when advanced to clinical trials, will help to distinguish those patients who are likely to benefit from a drug.