Deconstruction of Corticospinal Circuits for Goal-Directed Motor Skills.

Corticospinal neurons (CSNs) represent the direct cortical outputs to the spinal cord and play important roles in motor control across different species. However, their organizational principle remains unclear. By using a retrograde labeling system, we defined the requirement of CSNs in the execution of a skilled forelimb food-pellet retrieval task in mice. In vivo imaging of CSN activity during performance revealed the sequential activation of topographically ordered functional ensembles with moderate local mixing. Region-specific manipulations indicate that CSNs from caudal or rostral forelimb area control reaching or grasping, respectively, and both are required in the transitional pronation step. These region-specific CSNs terminate in different spinal levels and locations, therefore preferentially connecting with the premotor neurons of muscles engaged in different steps of the task. Together, our findings suggest that spatially defined groups of CSNs encode different movement modules, providing a logic for parallel-ordered corticospinal circuits to orchestrate multistep motor skills.

Enhanced compound-protein binding affinity prediction by representing protein multimodal information via a coevolutionary strategy.

Due to the lack of a method to efficiently represent the multimodal information of a protein, including its structure and sequence information, predicting compound-protein binding affinity (CPA) still suffers from low accuracy when applying machine-learning methods. To overcome this limitation, in a novel end-to-end architecture (named FeatNN), we develop a coevolutionary strategy to jointly represent the structure and sequence features of proteins and ultimately optimize the mathematical models for predicting CPA. Furthermore, from the perspective of data-driven approach, we proposed a rational method that can utilize both high- and low-quality databases to optimize the accuracy and generalization ability of FeatNN in CPA prediction tasks. Notably, we visually interpret the feature interaction process between sequence and structure in the rationally designed architecture. As a result, FeatNN considerably outperforms the state-of-the-art (SOTA) baseline in virtual drug evaluation tasks, indicating the feasibility of this approach for practical use. FeatNN provides an outstanding method for higher CPA prediction accuracy and better generalization ability by efficiently representing multimodal information of proteins via a coevolutionary strategy.

Image Encryption Using a New Hybrid Chaotic Map and Spiral Transformation.

Image encryption based on chaotic maps is an important method for ensuring the secure communication of digital multimedia on the Internet. To improve the encryption performance and security of image encryption systems, a new image encryption algorithm is proposed that employs a compound chaotic map and random cyclic shift. First, a new hybrid chaotic system is designed by coupling logistic, ICMIC, Tent, and Chebyshev (HLITC) maps. Comparison tests with previous chaotic maps in terms of chaotic trajectory, Lyapunov exponent, and approximate entropy illustrate that the new hybrid chaotic map has better chaotic performance. Then, the proposed HLITC chaotic system and spiral transformation are used to develop a new chaotic image encryption scheme using the double permutation strategy. The new HLITC chaotic system is used to generate key sequences used in the image scrambling and diffusion stages. The spiral transformation controlled by the chaotic sequence is used to scramble the pixels of the plaintext image, while the XOR operation based on a chaotic map is used for pixel diffusion. Extensive experiments on statistical analysis, key sensitivity, and key space analysis were conducted. Experimental results show that the proposed encryption scheme has good robustness against brute-force attacks, statistical attacks, and differential attacks and is more effective than many existing chaotic image encryption algorithms.