Eric Davidson's Regulatory Genome for Computer Science: Causality, Logic, and Proof Principles of the Genomic cis-Regulatory Code

Abstract

In this article, we discuss several computer science problems, inspired by our 15-year-long collaboration with Prof. Eric Davidson, focusing on computer science contributions to the study of the regulatory genome. Our joint study was inspired by his lifetime trailblazing research program rooted in causal gene regulatory networks (GRNs), system completeness, genomic Boolean logic, and genomically encoded regulatory information. We present first four inspiring questions that Eric Davidson asked, and the follow-up, namely, seven technical problems, fully or partially resolved with the methods of computer science. At the center, and unifying the intellectual backbone of those technical challenges, stands "Causality." Our collaboration produced the causalityinferred cisGRN-Lexicon database, containing the cis-regulatory architecture (CRA) of 600+ transcription factor (TF)-encoding genes and other regulatory genes, in eight species: human, mouse, fruit fly, sea urchin, nematode, rat, chicken, and zebrafish. These CRAs are causality- inferred regulatory regions of genes, derived experimentally through the experimental method called "cis-regulatory analysis" (also known as the "Davidson criteria"). In this research program, causality challenges for computer science show up in two components: (1) how to define data structures that represent the causality-inferred, by the Davidson criteria, DNA structure data and to define a versatile software system to host them; and (2) how to identify by automated software for text analysis the experimental technical articles applying the Davidson criteria to the analysis to genes. We next present the cisGRN-Lexicon Meta-Analysis (Part I). We conclude the article with some reflections on epistemology and philosophy themes concerning the role of causality, logic, and proof in the emerging elegant mathematical theory and practice of the regulatory genome. It is challenging to explain what "explanation" is, and to understand what "understanding" is, when the technical task is to "prove" system-level causality completeness of a 50-gene causal GRN. Within the Peter-Davidson Boolean GRN model, the Peter-Davidson completeness "theorem" provides a seminal answer: Experimental causality system completeness = Computational exact prediction completeness. The article is organized as follows. Section 2 is dedicated to our Prof. Eric Davidson. Section 3 gives a brief introduction for computer scientists to the regulatory genome and its information processing operations in terms similar to the electronic computer. Section 4 proposes to honor Eric Davidsons life-long scientific work on the regulatory genome by naming a most fundamental time unit constant after him. Section 5 presents four grand challenge questions that Eric Davidson asked, and seven follow-up problems inspired by the first two questions, which we fully or partially solved together. Central to the mentioned solutions is our construction of the cisGRN-Lexcion, the database of causally inferred CRA of 600+ regulatory genes in eight species. Section 6 presents Part I of the cisGRN-Lexcion Meta-Analysis, coached as "rules" of the genomic cis-regulatory code. Section 7 is devoted to reflections on epistemological and philosophical themes: causality, logic, and proof in the elegant mathematical modeling of the regulatory genome. We present here the "Davidsonian Causal Systems Biology Axioms," which guide us toward understanding of the meaning of "proving" causality completeness, for a complex experimental system, by exact computational predictions.