Regulatory effect of bacterial melanin on the isoforms of new superoxide-producing associates from rat tissues in rotenone-induced Parkinson's disease.

According to recent research, selective neuronal vulnerability in Parkinson's disease (PD) results from several phenotypic traits, including calcium-dependent, feed-forward control of mitochondrial respiration leading to elevated reactive oxygen species and cytosolic calcium concentration, an extensive axonal arbor, and a reactive neurotransmitter. Therefore, antioxidant therapy is a promising direction in the treatment of PD. In vitro studies have indicated the survival-promoting activity of bacterial melanin (BM) on midbrain dopaminergic neuron cultures. It has been established that BM has a number of protective and anti-inflammatory properties, so there is a high probability of a protective effect of BM in the early stages of PD. In this study, PD was induced through the unilateral intracerebral administration of rotenone followed by bacterial melanin. Tissues (brain, lungs, and small intestine) from the observed groups underwent isolation and purification to extract isoforms of new thermostable superoxide (О2-)-producing associates between NADPH-containing lipoprotein (NLP) and NADPH oxidase-Nox (NLP-Nox). The optical absorption spectral characteristics, specific amounts, stationary concentration of the produced О2-, and the content of NADPH in the observed associates were determined. The optical absorption spectra of the NLP-Nox isoforms in the visible and UV regions in the experimental groups did not differ from those of the control group. However, compared with the control group, the specific content of the total fractions of NLP-Nox isoforms associated with PD groups was higher, especially in the small intestine. These findings suggest that the described changes may represent a novel mechanism for rotenone-induced PD. Furthermore, bacterial melanin demonstrated antioxidant properties and regulated membrane formation in the brain, lung, and small intestine. This regulation occurred by inhibiting the release of new membrane-bound formations (NLP-Nox associates) from these membranes while simultaneously regulating the steady-state concentration of the formed О2-.

Plasticity in Motoneurons Following Spinal Cord Injury in Fructose-induced Diabetic Rats.

Spinal cord injury (SCI) causes motor impairment and the proper excitation/inhibition balance in motoneurons is important for recovery. Diabetes mellitus impairs regenerative capacity following SCI. The purpose of this study was to assess the short-term plasticity (STP) of lumbar spinal cord motoneurons in conditions of (1) lateral hemisection (SCI), (2) fructose-induced diabetes (D), and (3) diabetes associated with hemisection (D + SCI). We show that in the cases of SCI, D, and D + SCI, the ratio of percentage share of excitatory and inhibitory combinations of motoneurons responses to high-frequence stimulation of sciatic nerve is multidirectional. In the SCI and D + SCI groups, the cumulative changes in generalized baseline frequencies decreased significantly. When we compared the cumulative changes in the intensity of excitatory and inhibitory responses relative to baseline during high-frequency stimulation (tetanization epoch), we found that there was a significant intensification in tetanic depression in the D + SCI groups versus SCI, as well as an intensification in tetanic potentiation in the D + SCI vs. D and D + SCI vs. SCI groups. Thus, in conditions of traumatic and/or metabolic pathology, the distinct synaptic inputs exhibit opposing plasticity for homeostatic control of neurotransmission and these integral changes most likely shape postsynaptic STP in the spinal motor network.

Skilful precipitation nowcasting using deep generative models of radar.

Precipitation nowcasting, the high-resolution forecasting of precipitation up to two hours ahead, supports the real-world socioeconomic needs of many sectors reliant on weather-dependent decision-making1,2. State-of-the-art operational nowcasting methods typically advect precipitation fields with radar-based wind estimates, and struggle to capture important non-linear events such as convective initiations3,4. Recently introduced deep learning methods use radar to directly predict future rain rates, free of physical constraints5,6. While they accurately predict low-intensity rainfall, their operational utility is limited because their lack of constraints produces blurry nowcasts at longer lead times, yielding poor performance on rarer medium-to-heavy rain events. Here we present a deep generative model for the probabilistic nowcasting of precipitation from radar that addresses these challenges. Using statistical, economic and cognitive measures, we show that our method provides improved forecast quality, forecast consistency and forecast value. Our model produces realistic and spatiotemporally consistent predictions over regions up to 1,536 km × 1,280 km and with lead times from 5-90 min ahead. Using a systematic evaluation by more than 50 expert meteorologists, we show that our generative model ranked first for its accuracy and usefulness in 89% of cases against two competitive methods. When verified quantitatively, these nowcasts are skillful without resorting to blurring. We show that generative nowcasting can provide probabilistic predictions that improve forecast value and support operational utility, and at resolutions and lead times where alternative methods struggle.

Mastering Atari, Go, chess and shogi by planning with a learned model.

Constructing agents with planning capabilities has long been one of the main challenges in the pursuit of artificial intelligence. Tree-based planning methods have enjoyed huge success in challenging domains, such as chess1 and Go2, where a perfect simulator is available. However, in real-world problems, the dynamics governing the environment are often complex and unknown. Here we present the MuZero algorithm, which, by combining a tree-based search with a learned model, achieves superhuman performance in a range of challenging and visually complex domains, without any knowledge of their underlying dynamics. The MuZero algorithm learns an iterable model that produces predictions relevant to planning: the action-selection policy, the value function and the reward. When evaluated on 57 different Atari games3-the canonical video game environment for testing artificial intelligence techniques, in which model-based planning approaches have historically struggled4-the MuZero algorithm achieved state-of-the-art performance. When evaluated on Go, chess and shogi-canonical environments for high-performance planning-the MuZero algorithm matched, without any knowledge of the game dynamics, the superhuman performance of the AlphaZero algorithm5 that was supplied with the rules of the game.

Disruption of Cholinergic Circuits as an Area for Targeted Drug Treatment of Alzheimer's Disease: In Vivo Assessment of Short-Term Plasticity in Rat Brain.

The search for new therapeutics for the treatment of Alzheimer's disease (AD) is still in progress. Aberrant pathways of synaptic transmission in basal forebrain cholinergic neural circuits are thought to be associated with the progression of AD. However, the effect of amyloid-beta (Aß) on short-term plasticity (STP) of cholinergic circuits in the nucleus basalis magnocellularis (NBM) is largely unknown. STP assessment in rat brain cholinergic circuitry may indicate a new target for AD cholinergic therapeutics. Thus, we aimed to study in vivo electrophysiological patterns of synaptic activity in NBM-hippocampus and NBM-basolateral amygdala circuits associated with AD-like neurodegeneration. The extracellular single-unit recordings of responses from the hippocampal and basolateral amygdala neurons to high-frequency stimulation (HFS) of the NBM were performed after intracerebroventricular injection of Aß 25-35. We found that after Aß 25-35 exposure the number of hippocampal neurons exhibiting inhibitory responses to HFS of NBM is decreased. The reverse tendency was seen in the basolateral amygdala inhibitory neural populations, whereas the number of amygdala neurons with excitatory responses decreased. The low intensity of inhibitory and excitatory responses during HFS and post-stimulus period is probably due to the anomalous basal synaptic transmission and excitability of hippocampal and amygdala neurons. These functional changes were accompanied by structural alteration of hippocampal, amygdala, and NBM neurons. We have thus demonstrated that Aß 25-35 induces STP disruption in NBM-hippocampus and NBM-basolateral amygdala circuits as manifested by unbalanced excitatory/inhibitory responses and their frequency. The results of this study may contribute to a better understanding of synaptic integrity. We believe that advancing our understanding of in vivo mechanisms of synaptic plasticity disruption in specific neural circuits could lead to effective drug searches for AD treatment.

Improved protein structure prediction using potentials from deep learning.

Protein structure prediction can be used to determine the three-dimensional shape of a protein from its amino acid sequence1. This problem is of fundamental importance as the structure of a protein largely determines its function2; however, protein structures can be difficult to determine experimentally. Considerable progress has recently been made by leveraging genetic information. It is possible to infer which amino acid residues are in contact by analysing covariation in homologous sequences, which aids in the prediction of protein structures3. Here we show that we can train a neural network to make accurate predictions of the distances between pairs of residues, which convey more information about the structure than contact predictions. Using this information, we construct a potential of mean force4 that can accurately describe the shape of a protein. We find that the resulting potential can be optimized by a simple gradient descent algorithm to generate structures without complex sampling procedures. The resulting system, named AlphaFold, achieves high accuracy, even for sequences with fewer homologous sequences. In the recent Critical Assessment of Protein Structure Prediction5 (CASP13)-a blind assessment of the state of the field-AlphaFold created high-accuracy structures (with template modelling (TM) scores6 of 0.7 or higher) for 24 out of 43 free modelling domains, whereas the next best method, which used sampling and contact information, achieved such accuracy for only 14 out of 43 domains. AlphaFold represents a considerable advance in protein-structure prediction. We expect this increased accuracy to enable insights into the function and malfunction of proteins, especially in cases for which no structures for homologous proteins have been experimentally determined7.

Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13).

We describe AlphaFold, the protein structure prediction system that was entered by the group A7D in CASP13. Submissions were made by three free-modeling (FM) methods which combine the predictions of three neural networks. All three systems were guided by predictions of distances between pairs of residues produced by a neural network. Two systems assembled fragments produced by a generative neural network, one using scores from a network trained to regress GDT_TS. The third system shows that simple gradient descent on a properly constructed potential is able to perform on par with more expensive traditional search techniques and without requiring domain segmentation. In the CASP13 FM assessors' ranking by summed z-scores, this system scored highest with 68.3 vs 48.2 for the next closest group (an average GDT_TS of 61.4). The system produced high-accuracy structures (with GDT_TS scores of 70 or higher) for 11 out of 43 FM domains. Despite not explicitly using template information, the results in the template category were comparable to the best performing template-based methods.

A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play.

The game of chess is the longest-studied domain in the history of artificial intelligence. The strongest programs are based on a combination of sophisticated search techniques, domain-specific adaptations, and handcrafted evaluation functions that have been refined by human experts over several decades. By contrast, the AlphaGo Zero program recently achieved superhuman performance in the game of Go by reinforcement learning from self-play. In this paper, we generalize this approach into a single AlphaZero algorithm that can achieve superhuman performance in many challenging games. Starting from random play and given no domain knowledge except the game rules, AlphaZero convincingly defeated a world champion program in the games of chess and shogi (Japanese chess), as well as Go.

Mastering the game of Go without human knowledge.

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves using deep neural networks. These neural networks were trained by supervised learning from human expert moves, and by reinforcement learning from self-play. Here we introduce an algorithm based solely on reinforcement learning, without human data, guidance or domain knowledge beyond game rules. AlphaGo becomes its own teacher: a neural network is trained to predict AlphaGo's own move selections and also the winner of AlphaGo's games. This neural network improves the strength of the tree search, resulting in higher quality move selection and stronger self-play in the next iteration. Starting tabula rasa, our new program AlphaGo Zero achieved superhuman performance, winning 100-0 against the previously published, champion-defeating AlphaGo.

Learning Local Feature Descriptors Using Convex Optimisation.

The objective of this work is to learn descriptors suitable for the sparse feature detectors used in viewpoint invariant matching. We make a number of novel contributions towards this goal. First, it is shown that learning the pooling regions for the descriptor can be formulated as a convex optimisation problem selecting the regions using sparsity. Second, it is shown that descriptor dimensionality reduction can also be formulated as a convex optimisation problem, using Mahalanobis matrix nuclear norm regularisation. Both formulations are based on discriminative large margin learning constraints. As the third contribution, we evaluate the performance of the compressed descriptors, obtained from the learnt real-valued descriptors by binarisation. Finally, we propose an extension of our learning formulations to a weakly supervised case, which allows us to learn the descriptors from unannotated image collections. It is demonstrated that the new learning methods improve over the state of the art in descriptor learning on the annotated local patches data set of Brown et al. and unannotated photo collections of Philbin et al.

Immediate structured visual search for medical images.

The objective of this work is a scalable, real-time visual search engine for medical images. In contrast to existing systems that retrieve images that are globally similar to a query image, we enable the user to select a query Region Of Interest (ROI) and automatically detect the corresponding regions within all returned images. This allows the returned images to be ranked on the content of the ROI, rather than the entire image. Our contribution is two-fold: (i) immediate retrieval - the data is appropriately pre-processed so that the search engine returns results in real-time for any query image and ROI; (ii) structured output - returning ROIs with a choice of ranking functions. The retrieval performance is assessed on a number of annotated queries for images from the IRMA X-ray dataset and compared to a baseline.