Insulin-like peptides (ILPs) play highly conserved roles in development and physiology. Most animal genomes encode multiple ILPs. Here we identify mechanisms for how the forty Caenorhabditis elegans ILPs coordinate diverse processes, including development, reproduction, longevity and several specific stress responses. Our systematic studies identify an ILP-based combinatorial code for these phenotypes characterized by substantial functional specificity and diversity rather than global redundancy. Notably, we show that ILPs regulate each other transcriptionally, uncovering an ILP-to-ILP regulatory network that underlies the combinatorial phenotypic coding by the ILP family. Extensive analyses of genetic interactions among ILPs reveal how their signals are integrated. A combined analysis of these functional and regulatory ILP interactions identifies local genetic circuits that act in parallel and interact by crosstalk, feedback and compensation. This organization provides emergent mechanisms for phenotypic specificity and graded regulation for the combinatorial phenotypic coding we observe. Our findings also provide insights into how large hormonal networks regulate diverse traits.

There is a body of literature that describes the geometry and the physics of filopodia using either stochastic models or partial differential equations and elasticity and coarse-grained theory. Comparatively, there is a paucity of models focusing on the regulation of the network of proteins that control the formation of different actin structures. Using a combination of in-vivo and in-vitro experiments together with a system of ordinary differential equations, we focused on a small number of well-characterized, interacting molecules involved in actin-dependent filopodia formation: the actin remodeler Eps8, whose capping and bundling activities are a function of its ligands, Abi-1 and IRSp53, respectively; VASP and Capping Protein (CP), which exert antagonistic functions in controlling filament elongation. The model emphasizes the essential role of complexes that contain the membrane deforming protein IRSp53, in the process of filopodia initiation. This model accurately accounted for all observations, including a seemingly paradoxical result whereby genetic removal of Eps8 reduced filopodia in HeLa, but increased them in hippocampal neurons, and generated quantitative predictions, which were experimentally verified. The model further permitted us to explain how filopodia are generated in different cellular contexts, depending on the dynamic interaction established by Eps8, IRSp53 and VASP with actin filaments, thus revealing an unexpected plasticity of the signaling network that governs the multifunctional activities of its components in the formation of filopodia.

In this paper, we propose a novel application of Generative Adversarial Networks (GAN) to the synthesis of cells imaged by fluorescence microscopy. Compared to natural images, cells tend to have a simpler and more geometric global structure that facilitates image generation. However, the correlation between the spatial pattern of different fluorescent proteins reflects important biological functions, and synthesized images have to capture these relationships to be relevant for biological applications. We adapt GANs to the task at hand and propose new models with casual dependencies between image channels that can generate multichannel images, which would be impossible to obtain experimentally. We evaluate our approach using two independent techniques and compare it against sensible baselines. Finally, we demonstrate that by interpolating across the latent space we can mimic the known changes in protein localization that occur through time during the cell cycle, allowing us to predict temporal evolution from static images.