RESUMO
The high-dimensional character of most biological systems presents genuine challenges for modeling and prediction. Here we propose a neural network-based approach for dimensionality reduction and analysis of biological gene expression data, using, as a case study, a well-known genetic network in the early Drosophila embryo, the gap gene patterning system. We build an autoencoder compressing the dynamics of spatial gap gene expression into a two-dimensional (2D) latent map. The resulting 2D dynamics suggests an almost linear model, with a small bare set of essential interactions. Maternally defined spatial modes control gap genes positioning, without the classically assumed intricate set of repressive gap gene interactions. This, surprisingly, predicts minimal changes of neighboring gap domains when knocking out gap genes, consistent with previous observations. Latent space geometries in maternal mutants are also consistent with the existence of such spatial modes. Finally, we show how positional information is well defined and interpretable as a polar angle in latent space. Our work illustrates how optimization of small neural networks on medium-sized biological datasets is sufficiently informative to capture essential underlying mechanisms of network function.
Assuntos
Proteínas de Drosophila , Redes Reguladoras de Genes , Redes Neurais de Computação , Animais , Drosophila/embriologia , Drosophila/genética , Proteínas de Drosophila/genética , Proteínas de Drosophila/metabolismo , Modelos GenéticosRESUMO
A fundamental question about biomolecular condensates is how distinct condensates can emerge from the interplay of different components. Here we present a minimal model of droplet differentiation where phase separated droplets demix into two types with different chemical modifications triggered by enzymatic reactions. We use numerical solutions to Cahn-Hilliard equations with chemical reactions and an effective droplet model to reveal the switchlike behavior. Our work shows how condensate identities in cells could result from competing enzymatic actions.
RESUMO
The leading nonlinear stress response in a periodically strained concentrated colloidal dispersion is studied experimentally and by theory. A thermosensitive microgel dispersion serves as well-characterized glass-forming model, where the stress response at the first higher harmonic frequency (3ω for strain at frequency ω) is investigated in the limit of small amplitude. The intrinsic nonlinearity at the third harmonic exhibits a scaling behavior which has a maximum in an intermediate frequency window and diverges when approaching the glass transition. It captures the (in-) stability of the transient elastic structure. Elastic stresses in-phase with the third power of the strain dominate the scaling. Our results qualitatively differ from previously derived scaling behavior in dielectric spectroscopy of supercooled molecular liquids. This might indicate a dependence of the nonlinear response on the symmetry of the external driving under time reversal.