The staggering diversity of the living world is a testament to the amount of variation available to the agency of natural selection. While it has been assumed that variation is entirely uniform and unbiased, recent work has challenged this notion. Evolutionary developmental biology seeks to understand the biases on variation imposed by developmental processes and their distinction from selective constraints. Metazoan development is best described by developmental gene pathways which are composed of transcription factors, signaling molecules, and terminal differentiation genes. A systematic comparison of such pathways across species would reveal the patterns of conservation and divergence; however this has not yet been achieved. In the EvoDevoPaths project we will develop a new approach to unravel pathways using both single-cell and tissue-specific transcriptomics. Our aim is to elucidate the evolution of developmental gene pathways using intricate embryology in the nematode phylum, a single-cell transcriptomic method we have developed, and sophisticated computational approaches for pathway comparisons. We will ask how variation is distributed across the specification and differentiation modules of a pathway using the nematode endoderm pathway as a model system. We further propose that the evolutionary change in the tissue specification pathways of early cell lineages is constrained by the properties of cell specification pathways. To test this hypothesis we will, for the first time, determine early developmental cell lineages from single cell transcriptomic data. Finally, we will attempt to unify the molecular signatures of conserved stages in disparate phyla under a framework in which they can be systematically compared. This research collectively represents the first time that developmental gene pathways are examined in an unbiased manner contributing to a theory of molecular variation that explains the evolutionary processes that underlie phenotypic novelty.