LMFFNet: A Well-Balanced Lightweight Network for Fast and Accurate Semantic Segmentation.

Real-time semantic segmentation is widely used in autonomous driving and robotics. Most previous networks achieved great accuracy based on a complicated model involving mass computing. The existing lightweight networks generally reduce the parameter sizes by sacrificing the segmentation accuracy. It is critical to balance the parameters and accuracy for real-time semantic segmentation. In this article, we propose a lightweight multiscale-feature-fusion network (LMFFNet) mainly composed of three types of components: split-extract-merge bottleneck (SEM-B) block, feature fusion module (FFM), and multiscale attention decoder (MAD), where the SEM-B block extracts sufficient features with fewer parameters. FFMs fuse multiscale semantic features to effectively improve the segmentation accuracy and the MAD well recovers the details of the input images through the attention mechanism. Without pretraining, LMFFNet-3-8 achieves 75.1% mean intersection over union (mIoU) with 1.4 M parameters at 118.9 frames/s using RTX 3090 GPU. More experiments are investigated extensively on various resolutions on other three datasets of CamVid, KITTI, and WildDash2. The experiments verify that the proposed LMFFNet model makes a decent tradeoff between segmentation accuracy and inference speed for real-time tasks. The source code is publicly available at https://github.com/Greak-1124/LMFFNet.

Probing the catalytic center of porcine aminoacylase 1 by site-directed mutagenesis, homology modeling and substrate docking.

Three-dimensional structural models of porcine aminoacylase 1 (pACY1) were constructed by homology modeling and aligning the structures of members of the M20 peptidase family. After energy minimization and quality evaluation, the best model from the homology modeling was chosen for docking with the best substrate (N-acetyl-L-methionine). The most reasonable binding mode was found among a large number of predicted complexes by using clustering analysis and screening with expert knowledge. Structural analysis revealed that the zinc ion is not likely to bind to the substrate, and that Arg348 and Glu146 play vital roles in binding and catalysis. In the site-directed mutagenesis experiments, mutation of His79, Asp112, Glu147, Arg348, and Glu146, resulted in significant reductions of specific activity, while the wild-type pACY1 overexpressed in Rosetta (DE3) had almost as high a specific activity as the native enzyme. On the basis of these observations, we proposed a revised catalytic mechanism for this metalloenzyme.