Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, the quality of public data might be different than that of industry data due to different labs reporting measurements, different measurement techniques, fewer samples and less diverse and specialized assays. As part of a European funded project (ExCAPE), that brought together expertise from pharmaceutical industry, machine learning, and high-performance computing, we investigated how well machine learning models obtained from public data can be transferred to internal pharmaceutical industry data. Our results show that machine learning models trained on public data can indeed maintain their predictive power to a large degree when applied to industry data. Moreover, we observed that deep learning derived machine learning models outperformed comparable models, which were trained by other machine learning algorithms, when applied to internal pharmaceutical company datasets. To our knowledge, this is the first large-scale study evaluating the potential of machine learning and especially deep learning directly at the level of industry-scale settings and moreover investigating the transferability of publicly learned target prediction models towards industrial bioactivity prediction pipelines.
Recombinant proteins play pivotal roles in numerous applications including industrial biocatalysts or therapeutics. Despite the recent progress in computational protein structure prediction, protein solubility and reduced aggregation propensity remain challenging attributes to design. Identification of aggregation-prone regions is essential for understanding misfolding diseases or designing efficient protein-based technologies, and as such has a great socio-economic impact. Here, we introduce AggreProt, a user-friendly webserver that automatically exploits an ensemble of deep neural networks to predict aggregation-prone regions (APRs) in protein sequences. Trained on experimentally evaluated hexapeptides, AggreProt compares to or outperforms state-of-the-art algorithms on two independent benchmark datasets. The server provides per-residue aggregation profiles along with information on solvent accessibility and transmembrane propensity within an intuitive interface with interactive sequence and structure viewers for comprehensive analysis. We demonstrate AggreProt efficacy in predicting differential aggregation behaviours in proteins on several use cases, which emphasize its potential for guiding protein engineering strategies towards decreased aggregation propensity and improved solubility. The webserver is freely available and accessible at https://loschmidt.chemi.muni.cz/aggreprot/.
The ANTAREX project aims at expressing the application self-adaptivity through a Domain Specific Language (DSL) and to runtime manage and autotune applications for green and heterogeneous High Performance Computing (HPC) systems up to Exascale. The DSL approach allows the definition of energy-efficiency, performance, and adaptivity strategies as well as their enforcement at runtime through application autotuning and resource and power management. We show through a mini-app extracted from one of the project application use cases some initial exploration of application precision tuning by means enabled by the DSL.
The examination of new (3 + 1)-dimensional wave equations is undertaken in this study. Initially, the identification of the Lie symmetries of the model is carried out through the utilization of the Lie symmetry approach. The commutator and adjoint table of the symmetries are presented. Subsequently, the model under discussion is transformed into an ordinary differential equation using these symmetries. The construction of several bright, kink, and dark solitons for the suggested equation is then achieved through the utilization of the new auxiliary equation method. Subsequently, an analysis of the dynamical nature of the model is conducted, encompassing various angles such as bifurcation, chaos, and sensitivity. Bifurcation occurs at critical points within a dynamical system, accompanied by the application of an outward force, which unveils the emergence of chaotic phenomena. Two-dimensional plots, time plots, multistability, and Lyapunov exponents are presented to illustrate these chaotic behaviors. Furthermore, the sensitivity of the investigated model is executed utilizing the Runge–Kutta method. This analysis confirms that the stability of the solution is minimally affected by small changes in initial conditions. The attained outcomes show the effectiveness of the presented methods in evaluating solitons of multiple nonlinear models.
The Sentinel-1 satellite system continuously observes European countries at a relatively high revisit frequency of six days per orbital track. Given the Sentinel-1 configuration, most areas in Czechia are observed every 1–2 days by different tracks in a moderate resolution. This is attractive for various types of analyses by various research groups. The starting point for interferometric (InSAR) processing is an original data provided in a Single Look Complex (SLC) level. This work represents advantages of storing data augmented to a specifically corrected level of data, SLC-C. The presented database contains Czech nationwide Sentinel-1 data stored in burst units that have been pre-processed to the state of a consistent well-coregistered dataset of SLC-C. These are resampled SLC data with their phase values reduced by a topographic phase signature, ready for fast interferometric analyses (an interferogram is generated by a complex conjugate between two stored SLC-C files). The data can be used directly into multitemporal interferometry techniques, e.g., Persistent Scatterers (PS) or Small Baseline (SB) techniques applied here. A further development of the nationwide system utilising SLC-C data would lead into a dynamic state where every new pre-processed burst triggers a processing update to detect unexpected changes from InSAR time series and therefore provides a signal for early warning against a potential dangerous displacement, e.g., a landslide, instability of an engineering structure or a formation of a sinkhole. An update of the processing chain would also allow use of cross-polarised Sentinel-1 data, needed for polarimetric analyses. The current system is running at a national supercomputing centre IT4Innovations in interconnection to the Czech Copernicus Collaborative Ground Segment (CESNET), providing fast on-demand InSAR results over Czech territories. A full nationwide PS processing using data over Czechia was performed in 2017, discovering several areas of land deformation. Its downsampled version and basic findings are demonstrated within the article.
Remote-sensing-driven urban change detection has been studied in many ways for decades for a wide field of applications, such as understanding socio-economic impacts, identifying new settlements, or analyzing trends of urban sprawl. Such kinds of analyses are usually carried out manually by selecting high-quality samples that binds them to small-scale scenarios, either temporarily limited or with low spatial or temporal resolution. We propose a fully automated method that uses a large amount of available remote sensing observations for a selected period without the need to manually select samples. This enables continuous urban monitoring in a fully automated process. Furthermore, we combine multispectral optical and synthetic aperture radar (SAR) data from two eras as two mission pairs with synthetic labeling to train a neural network for detecting urban changes and activities. As pairs, we consider European Remote Sensing (ERS-1/2) and Landsat 5 Thematic Mapper (TM) for 1991–2011 and Sentinel 1 and 2 for 2017–2021. For every era, we use three different urban sites—Limassol, Rotterdam, and Liège—with at least each, and deep observation time series with hundreds and up to over a thousand of samples. These sites were selected to represent different challenges in training a common neural network due to atmospheric effects, different geographies, and observation coverage. We train one model for each of the two eras using synthetic but noisy labels, which are created automatically by combining state-of-the-art methods, without the availability of existing ground truth data. To combine the benefit of both remote sensing types, the network models are ensembles of optical- and SAR-specialized sub-networks. We study the sensitivity of urban and impervious changes and the contribution of optical and SAR data to the overall solution. Our implementation and trained models are available publicly to enable others to utilize fully automated continuous urban monitoring.