<span lang="EN-US">This article discusses a large number of textural features and integral transformations for the analysis of texture-type images. It also discusses the description and analysis of the features of applying existing methods for segmenting texture areas in images and determining the advantages and disadvantages of these methods and the problems that arise in the segmentation of texture areas in images. The purpose of the ongoing research is to use methods and determine the effectiveness of methods for the analysis of aerospace images, which are a combination of textural regions of natural origin and artificial objects. Currently, the automation of the processing of aerospace information, in particular images of the earth’s surface, remains an urgent task. The main goal is to develop models and methods for more efficient use of information technologies for the analysis of multispectral texture-type images in the developed algorithms. The article proposes a comprehensive approach to these issues, that is, the consideration of a large number of textural features by integral transformation to eventually create algorithms and programs applicable to solving a wide class of problems in agriculture.</span><p> </p>

According to WHO, the number of people with disabilities in the world has exceeded 1 billion. At the same time, 80 percent of all people with disabilities live in developing countries. In this regard, the demand for the use of applications for people with disabilities is growing every day. The paper deals with neural network methods like MediaPipe Holistic and the LSTM module for determining the sign language of people with disabilities. MediaPipe has demonstrated unprecedented low latency and high tracking accuracy in real-world scenarios thanks to built-in monitoring solutions. Therefore, MediaPipe Holistic was used in this work, which combines pose, hand, and face control with detailed levels. The main purpose of this paper is to show the effectiveness of the HAR algorithm for recognizing human actions, based on the architecture of in-depth learning for classifying actions into seven different classes.  The main problem of this paper is the high level of recognition of the sign language of people with disabilities when implementing their work in cross-platform applications, web applications, social networks that facilitate the daily life of people with disabilities and interact with society. To solve this problem, an algorithm was used that combines the architecture of a convolutional neural network (CNN) and long short-term memory (LSTM) to study spatial and temporal capabilities from three-dimensional skeletal data taken only from a Microsoft Kinect camera. This combination allows you to take advantage of LSTM when modeling temporal data and CNN when modeling spatial data. The results obtained based on calculations carried out by adding a new layer to the existing model showed higher accuracy than calculations carried out on the existing model.

Objectives: During the last two decades, HIV-1 has been spreading rapidly in former Soviet Union republics including Kyrgyzstan. The current molecular monitoring of HIV-infection epidemic is carried out in Russia only with no or limited data from the other FSU countries. The aim of this work was to investigate the prevalence of HIV-1 genetic variants circulating in Kyrgyzstan. Methods: Blood collection from the HIV-infected patients was carried out by local specialists with the informed consent and the questionnaire was answered by each of the patients. The total number of samples was 100. The washed cell pellets were transferred to Moscow following with proviral DNA extraction, PCR amplification and gag, pol and env genes sequencing. The phylogenetic analysis of nucleotide sequences using neighbor-joining method was carried out by MEGA 3 program. The preliminary data were obtained in 22 samples isolated from PBMC of HIV-infected patients from Kyrgyzstan. Results: Among the samples studied 6 (27.3%) samples belonged to a subtype CRF02_AG, 16 samples - to subtype A (A1). One of the samples belonging to CRF02_AG, probably, is a recombinant between CRF02_AG and A1. There was no major drug resistance mutations in the samples studied. The minor mutations were presented in small proportions: 1 in PR (L10I), 6 in RT (A62V - in 3 samples, V108G, E138A, Y181F, M184I, L210M - on one sample) and 1 in IN (L74M). It was impossible to associate the distribution of mutations with HIV-1 genetic variant. The V3 loop (env gene) in 17 samples was analyzed for tropism using geno2pheno program; all samples were found to be R5-viruses. Conclusion: The HIV-1 subtype A seems to dominate in Kyrgyzstan like in other FSU countries. The recombinant CRF02_AG epidemiologically linked to Uzbekistan is quite widespread. The rest of Kyrgyzstan collection is under investigation and the data will be refined soon

In this work, we present the draft genome sequence of Komagataeibacter europaeus strain GH1, which is an extremely efficient biocellulose producer.

Predicting the function of proteins is a crucial part of genome annotation, which can help in solving a wide range of biological problems. Many methods are available to predict the functions of proteins. However, except for sequence, most features are difficult to obtain or are not available for many proteins, which limits their scope. In addition, the performance of sequence-based feature prediction methods is often lower than that of methods that involve multiple features, and protein feature prediction can be time-consuming. Recent advances in this field are associated with the development of machine learning, which shows great progress in solving the problem of predicting protein functions. Today, however, most protein sequences have the status of «uncharacterized» or «putative». The need to assess the accuracy of identification of protein functions is an urgent task for machine learning approaches used to predict protein functions. In this study, the performance of two popular function prediction algorithms (ProtCNN and BiLSTM) was assessed from two perspectives and the procedures for building these models were described. As a result of the study of Pfam families, ProtCNN achieves an accuracy rate of 0.988 % and bidirectional LSTM has an accuracy rate of 0.9506 %. The use of the Pfam dataset allowed increasing the classification accuracy due to the large training dataset. The quality of the prediction increases with a large amount of training data. The study demonstrated that machine learning algorithms can be used as an effective tool for building protein function prediction models, in particular, the CNN network can be adapted as an accurate tool for annotating protein functions in the presence of large datasets.