Search for: All records

Abstract MicroRNAs (miRNAs) are important regulators of genes expression. Their levels are precisely controlled through modulating the activity of the microprocesser complex (MC). Here, we report that JANUS, a homology of the conserved U2 snRNP assembly factor in yeast and human, is required for miRNA accumulation. JANUS associates with MC components Dicer-like 1 (DCL1) and SERRATE (SE) and directly binds the stem-loop of pri-miRNAs. In a hypomorphic janus mutant, the activity of DCL1, the numbers of MC, and the interaction of primary miRNA transcript (pri-miRNAs) with MC are reduced. These data suggest that JANUS promotes the assembly and activity of MC through its interaction with MC and/or pri-miRNAs. In addition, JANUS modulates the transcription of some pri-miRNAs as it binds the promoter of pri-miRNAs and facilitates Pol II occupancy of at their promoters. Moreover, global splicing defects are detected in janus. Taken together, our study reveals a novel role of a conserved splicing factor in miRNA biogenesis. 

n this article, we present a novel and flexible multitask multilayer Bayesian mapping framework with readily extendable attribute layers. The proposed framework goes beyond modern metric-semantic maps to provide even richer environmental information for robots in a single mapping formalism while exploiting intralayer and interlayer correlations. It removes the need for a robot to access and process information from many separate maps when performing a complex task, advancing the way robots interact with their environments. To this end, we design a multitask deep neural network with attention mechanisms as our front-end to provide heterogeneous observations for multiple map layers simultaneously. Our back-end runs a scalable closed-form Bayesian inference with only logarithmic time complexity. We apply the framework to build a dense robotic map, including metric-semantic occupancy and traversability layers. Traversability ground truth labels are automatically generated from exteroceptive sensory data in a self-supervised manner. We present extensive experimental results on publicly available datasets and data collected by a three-dimensional bipedal robot platform and show reliable mapping performance in different environments. Finally, we also discuss how the current framework can be extended to incorporate more information, such as friction, signal strength, temperature, and physical quantity concentration using Gaussian map layers. The software for reproducing the presented results or running on customized data is made publicly available.