Measures of gene indispensability reveal the parallelism between phylogeny and ontogeny

Haeckel's biogenetic law, or the recapitulation theory remains a controversial subject to this day since it is not based on quantitative regularities. Currently, the modern version of the biogenetic law is the hourglass model with its phylotypic period. Importantly, the hourglass model does not provide evidence that ontogeny recapitulates phylogeny, and it is often presented as the alternative model to recapitulation theory. However, the hourglass model and recapitulation are not mutually exclusive, and there are several examples of recapitulation-like processes observable after the phylotypic period of ontogeny. At the level of transcriptomics, all attempts to demonstrate recapitulation failed. Using a novel approach, combining transcriptomics with phylostratigraphy, we demonstrate that recapitulation exists as a quantitatively testable law. We show that the mean indispensability of genes decreases for phylogenetically younger genes, as well as genes expressed during later stages of ontogeny. We also define the ontotypic period of phylogeny, an analog to the phylotypic period of ontogeny. Since it starts from the beginning of phylogeny, it is reasonable to hypothesize that recapitulation starts from the phylotypic period. We conclude that parallelism, or recapitulation, is explainable by the fact that genes that emerged later in phylogeny have tendencies to be expressed during later stages of ontogeny. This is expected from the fundamental properties of Darwinian evolution which is based on conservation of existing gene functions by purifying selection and addition of new gene functions by positive selection. According to this, our result (the existence of parallelism or recapitulation) represents the empirical evidence for macroevolution.