One of challenging goals of developmental biology is understand of genetic inheritability of developmental programming. This challenging task first requires identification of all genetic factors and their functional dependences, revealing holistic view of genetic organization of cellular and organismal development. Because of high complexity of genotype as well as phenotype, genetic dissection of developmental programs could be even untouchable by combinatorial explosion of the number of possible associations. Therefore, modern genetics needs to be more systematic and predictive. Recently we proposed network-guided approach for genetics of complex traits. First, we construct probabilistic functional gene networks for cells or organisms by benchmarking and integrating heterogeneous multi-omics data that are in general publicly available. Then, using guilt-by-association, and other algorithms of network propagation of known biological information, we predict gene functions, phenotypic effect of loss-of-function, and epistatic interaction. The information can contribute to reconstruction of map between genotype and phenotype. The network-guided genetics method has been effectively applied for various organisms; from simple microbe yeast, to multicellular animal C. elegans, and to the human.