MOTIVATION: Protein-protein interactions drive wide-ranging molecular processes, and characterizing at the atomic level how proteins interact (beyond just the fact that they interact) can provide key insights into understanding and controlling this machinery. Unfortunately, experimental determination of three-dimensional protein complex structures remains difficult and does not scale to the increasingly large sets of proteins whose interactions are of interest. Computational methods are thus required to meet the demands of large-scale, high-throughput prediction of how proteins interact, but unfortunately, both physical modeling and machine learning methods suffer from poor precision and/or recall. RESULTS: In order to improve performance in predicting protein interaction interfaces, we leverage the best properties of both data- and physics-driven methods to develop a unified Geometric Deep Neural Network, 'PInet' (Protein Interface Network). PInet consumes pairs of point clouds encoding the structures of two partner proteins, in order to predict their structural regions mediating interaction. To make such predictions, PInet learns and utilizes models capturing both geometrical and physicochemical molecular surface complementarity. In application to a set of benchmarks, PInet simultaneously predicts the interface regions on both interacting proteins, achieving performance equivalent to or even much better than the state-of-the-art predictor for each dataset. Furthermore, since PInet is based on joint segmentation of a representation of a protein surfaces, its predictions are meaningful in terms of the underlying physical complementarity driving molecular recognition. AVAILABILITY AND IMPLEMENTATION: PInet scripts and models are available at https://github.com/FTD007/PInet. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

In this paper, a simple image processing unit (IPU) structure for a general-purpose representation and association machine system is proposed. We immediately apply our IPU to a digit recognition and hyper-acuity test. The quality of the digit images has been severely degraded to mimic visual information variability experienced in human visual systems. The degraded images are then fed to the IPU, which consists of a randomly constructed low-density parity check (LDPC) code, an iterative decoder, a switch, and scaling, and decision devices. The results show that: 1) our IPU can reliably recognize digits despite the image quality being poor; 2) our IPU provides a hyper-acuity capability comparable to human visual systems; and 3) our IPU with the randomly constructed LDPC code can provide significantly improved recognition capability compared with the IPU without coding, even though the code is not optimized for our tasks.

In this paper we show that a small amount of bit stream features can be used to accurately estimate the energy consumption of state-of-the-art software and hardware accelerated decoder implementations for four different video codecs. By testing the estimation performance on HEVC, H.264, H.263, and VP9 we show that the proposed model can be used for any hybrid video codec. We test our approach on a high amount of different test sequences to prove the general validity. We show that less than 20 features are sufficient to obtain mean estimation errors that are smaller than 8%. Finally, an example will show the performance trade-offs in terms of rate, distortion, and decoding energy for all tested codecs.