Integrating pathway knowledge with deep neural networks to reduce the dimensionality in single-cell RNA-seq data

Background Single-cell RNA sequencing (scRNA-seq) data provide valuable insights into cellular heterogeneity which is significantly improving the current knowledge on biology and human disease. One of the main applications of scRNA-seq data analysis is the identification of new cell types and cell states. Deep neural networks (DNNs) are among the best methods to address this problem. However, this performance comes with the trade-off for a lack of interpretability in the results. In this work we propose an intelligible pathway-driven neural network to correctly solve cell-type related problems at single-cell resolution while providing a biologically meaningful representation of the data. Results In this study, we explored the deep neural networks constrained by several types of prior biological information, e.g. signaling pathway information, as a way to reduce the dimensionality of the scRNA-seq data. We have tested the proposed biologically-based architectures on thousands of cells of human and mouse origin across a collection of public datasets in order to check the performance of the model. Specifically, we tested the architecture across different validation scenarios that try to mimic how unknown cell types are clustered by the DNN and how it correctly annotates cell types by querying a database in a retrieval problem. Moreover, our approach demonstrated to be comparable to other less interpretable DNN approaches constrained by using protein-protein interactions gene regulation data. Finally, we show how the latent structure learned by the network could be used to visualize and to interpret the composition of human single cell datasets. Conclusions Here we demonstrate how the integration of pathways, which convey fundamental information on functional relationships between genes, with DNNs, that provide an excellent classification framework, results in an excellent alternative to learn a biologically meaningful representation of scRNA-seq data. In addition, the introduction of prior biological knowledge in the DNN reduces the size of the network architecture. Comparative results demonstrate a superior performance of this approach with respect to other similar approaches. As an additional advantage, the use of pathways within the DNN structure enables easy interpretability of the results by connecting features to cell functionalities by means of the pathway nodes, as demonstrated with an example with human melanoma tumor cells.

Novel Greener Microwave-Assisted Deprotection Methodology for the 1,3-Dioxolane Ketal of Isatin Using Calix[n]arenes

In this work, the combined use of p-sulfonic acid-calix[n]arene and microwave energy to hydrolyze the 1,3-dioxolane ketal of isatin was evaluated with excellent results. This is the first time that p-sulfonic acid-calix[n]arene has been used as the catalyst in a ketal hydrolysis reaction and the deprotection of the ketone carbonyl of isatin is reported. The presence of 2.5 mol% of p-sulfonic acid-calix[4,6]arene at 160 ºC resulted in over 96% conversion of this ketal in 10 min, with the additional advantage of using water as a solvent. This catalytic system (aqueous phase containing p-sulfonic acid-calix[4]arene) was reused for five consecutive cycles, with a conversion above 96% maintained. This reusable system is not practicable using p-toluenesulfonic acid and p-hydroxybenzenesulfonic acid as catalysts since both are extracted to the organic phase with the reaction product. The hydrolysis of 1,3-dioxolane ketal of isatins with different substituents (CH3, I, Br, Cl, F, NO2) in the aromatic ring was also evaluated. The protecting group of 5-methyl-isatin was removed with 73% conversion using 2.5 mol% of p-sulfonic acid-calix[4]arene at 160 ºC for 5 min. In contrast, the ketal of 5-nitro-isatin reached 80% conversion using the same conditions after 40 min.

Advantages of “Soft Steroids” in the Treatment of Inflammatory Eye Diseases. Оverview

The purpose to assess the benefits of using soft steroids in the treatment of inflammatory eye disorders according to literature data.Methods. The literature review concerning the administration of the gluco-corticosteroids and combination drugs based on gluco-corticosteroid for the treatment of inflammatory eye disorders. Both Russian and foreign sources for the 1980–2021 period were analyzed.The results. The combination drugs containing anti-infective drugs and gluco-corticosteroids are actively applied for the treatment of inflammatory eye disorders. That exerts joint ethiopathogenetic effect on the disorder. However, gluco-corticosteroid being a part of such drugs (predominantly dexamethasone) as often as not leads to ocular hypertension. In order to deal with this problem the so-called soft steroids (also classified as gluco-corticosteroids) were introduced. They lessen the possibility of the ocular hypertension and are marked by high efficiency and increased safety profile. One of the representatives of soft steroids is fluorometholone. There is a large evidential base in the modern literature that confirms much lesser influence of fluorometholone on ocular pressure if compared to dexamethasone. At the same time, dexamethasone has a higher anti-inflammatory activity, while on the other hand, its systemic immunosuppressive activity is lower. What is more, in terms of influence on the ocular surfaces, dexamethasone has an additional advantage which is causing mucin expression in conjuctival and corneal epithelium. The above mentioned merits of dexamethasone served as basis for its inclusion into the combination drug called Floas-T which is essentially the combination of tobramycin 0.3 % and fluorometholone 0.1 %. It is used in the treatment of inflammatory eye diseases and diseases of eye appendages, as well as for profylaxis of the diseases in the postoperative period.Conclusion. Combination drugs containing anti-infective components and gluco-corticosteroids seem to be highly promising for the treatment of inflammatory eye diseases. One of them worth highlighting is Floas-T classified as soft steroids containing tobramycin and fluorometholone. It compares to dexamethasone favourably in terms of efficiency, while contributing less to ocular hypertension.

Expert-enhanced machine learning for cardiac arrhythmia classification

We propose a new method for the classification task of distinguishing atrial fibrillation (AFib) from regular atrial tachycardias including atrial flutter (AFlu) based on a surface electrocardiogram (ECG). Recently, many approaches for an automatic classification of cardiac arrhythmia were proposed and to our knowledge none of them can distinguish between these two. We discuss reasons why deep learning may not yield satisfactory results for this task. We generate new and clinically interpretable features using mathematical optimization for subsequent use within a machine learning (ML) model. These features are generated from the same input data by solving an additional regression problem with complicated combinatorial substructures. The resultant can be seen as a novel machine learning model that incorporates expert knowledge on the pathophysiology of atrial flutter. Our approach achieves an unprecedented accuracy of 82.84% and an area under the receiver operating characteristic (ROC) curve of 0.9, which classifies as “excellent” according to the classification indicator of diagnostic tests. One additional advantage of our approach is the inherent interpretability of the classification results. Our features give insight into a possibly occurring multilevel atrioventricular blocking mechanism, which may improve treatment decisions beyond the classification itself. Our research ideally complements existing textbook cardiac arrhythmia classification methods, which cannot provide a classification for the important case of AFib↔AFlu. The main contribution is the successful use of a novel mathematical model for multilevel atrioventricular block and optimization-driven inverse simulation to enhance machine learning for classification of the arguably most difficult cases in cardiac arrhythmia. A tailored Branch-and-Bound algorithm was implemented for the domain knowledge part, while standard algorithms such as Adam could be used for training.

Teaching Human-Computer Interaction Modules—And Then Came COVID-19

In teaching Human-Computer Interaction at university level, it has always been beneficial to explain the related theory and engage students in a practical way, whether individually or in groups. And then came COVID-19. Face-to-face classes were replaced by emergency remote teaching methods. Students became student numbers in cyber space. The danger became real to convert back to the traditional way of presenting lectures, namely a lecturer doing all the talking and the students being the passive audience. This paper describes how the author had to adapt and innovate in terms of teaching Human-Computer Interaction modules to university students in a practical way during the COVID-19 pandemic. Frequent online quizzes, audio messages, online group discussion, smaller topic-dedicated practical activities, and webinars encouraging student participation, were employed. Instead of having access to eye-tracking technology in a usability laboratory, students had to innovate for usability evaluation assignments by employing observation, think-aloud protocols, and performance and self-reported metrics as data gathering methods. The laboratory had to be replaced by COVID-compliant places of residence. The outcomes of adapting previously-used teaching methods and inventing new ways to encourage student participation, were surprisingly positive. An additional advantage was that many of these methods turned out to be so successful that their application could be continued and extended to post-pandemic times for a blended learning approach to further enrich Human-Computer Interaction teaching.

Aligning Patient’s Ideas of a Good Life with Medically Indicated Therapies in Geriatric Rehabilitation Using Smart Sensors

New technologies such as smart sensors improve rehabilitation processes and thereby increase older adults’ capabilities to participate in social life, leading to direct physical and mental health benefits. Wearable smart sensors for home use have the additional advantage of monitoring day-to-day activities and thereby identifying rehabilitation progress and needs. However, identifying and selecting rehabilitation priorities is ethically challenging because physicians, therapists, and caregivers may impose their own personal values leading to paternalism. Therefore, we develop a discussion template consisting of a series of adaptable questions for the patient–physician encounter based on the capability approach. The goal is to improve geriatric rehabilitation and thereby increase participation in social life and well-being. To achieve this goal, we first analyzed what is considered important for participation on basis of the capability approach, human rights, and ethics of care. Second, we conducted an ethical analysis of each of the four identified dimensions of participation: political, economic, socio-cultural, and care. To improve compliance with rehabilitation measures, health professionals must align rehabilitation measures in an open dialogue with the patient’s aspiration for participation in each dimension. A discussion template based on the capability approach allows for a proactive approach in patient information and stimulates a critical assessment of treatment alternatives while reducing the risk of imposing personal values.

Rehydration of freeze substituted brain tissue for preembedding immuno-electron microscopy

Electron microscopy (EM) volume reconstruction is a powerful tool for investigating the fundamental structure of brain circuits, but the full potential of this technique is limited by the difficulty of integrating molecular information. High quality ultrastructural preservation is necessary for EM reconstruction, and intact, highly contrasted cell membranes are essential for following small neuronal processes through serial sections. Unfortunately, the antibody labeling methods used to identify most endogenous molecules result in compromised morphology, especially of membranes. Cryofixation can produce superior morphological preservation and has the additional advantage of allowing indefinite storage of valuable samples. We have developed a method based on cryofixation that allows sensitive immunolabeling of endogenous molecules, preserves excellent ultrastructure, and is compatible with high-contrast staining for serial EM reconstruction.

Recent Modalities in the Diagnosis of Obesity

Obesity has been an emerging health problem worldwide which has a major impact on public health. It is associated with medical, psychosocial and economic implications with increasing prevalence among both adult and paediatric population. Obesity led to an increased risk of medical conditions like diabetes mellitus, hypertension, coronary artery disease, insulin resistance and sleep apnoea. Obesity has a major impact on cardiovascular system causing structural and functional changes leading to cardiac dysfunction. Hence it is important to diagnose obesity at the earliest for timely prevention of associated complications. Apart from routine diagnostic methods for obesity like body mass index, anthropometry (waist circumference, hip circumference, neck circumference), several recent modalities were described for the diagnosis of obesity like radioimaging, nuclear medicine imaging which will be described in detail in this review article. MRI (Magnetic resonance imaging) aids in the detection of adipose tissue at various sites and organs, whereas MRS (Magnetic resonance spectroscopy) helps in mapping of small quantity of lipids. MRI helps in delineating ectopic adipose tissue accumulation establishing that obesity alone is not a major cause for derangement in metabolic profile. An additional advantage is MRI brain is an excellent imaging guide for studying the role of central appetite regulatory systems in the occurrence of obesity. Sonography is not accurate in the estimation of hepatic steatosis. But advancements in sonographic modalities gives an extra edge in evaluation of hepatic steatosis by availing special physical characteristics such as stiffness of adipose tissue and its sound absorption. Positron Emission Tomography (PET) (Nuclear medicine imaging) helps in studying central pathophysiology, activity of brown adipose tissue and disruption of gut-brain homeostasis.

Modified Regeneration Scheme for Energy Efficient Gas Dehydration

Molecular Sieve Dehydration units are used for dehydration of natural gas prior to gas processing or transportation. A molecular sieve dehydration system consists of multiple adsorbers which remove water during adsorption cycle until they get saturated with water. Regeneration of a saturated adsorber is performed by passing a hot regeneration gas stream through the adsorber. The hot regeneration gas after passing though the adsorber is then cooled before sending to regeneration gas compression. If an aircooled exchanger is used to cool the hot regeneration gas, heat available in the hot spent regeneration gas ends up in the atmosphere. In this context, an in-house study was performed to examine techno-economic viability of waste heat recovery from the hot spent regeneration gas using a modified regeneration scheme at one of the gas processing sites. The modified scheme involves installation of a new waste heat recovery (WHR) exchanger to exchange the heat available in the hot regeneration gas with regeneration heater's inlet regeneration gas thereby reducing the fuel gas consumption in the regeneration heater as well as power consumption in regeneration gas cooler fans. The study comprised design and operation data collection and analysis followed by assessment of key challenges. The key challenges include performance of the heater in WHR case (i.e. lower fuel gas consumption), space availability for the new WHR exchanger and modifications in the existing system. A thermodynamic model was developed for running various operating scenarios and estimating the WHR potential, including heater's specific fuel gas consumption analysis at varying temperatures, to establish realistic fuel gas savings. Overall, the study has indicated significant energy savings with good financial indicators for the proposed regeneration scheme. It has also showed reduction of peak heat duty of heater & cooler, thus providing an additional advantage of reduced CAPEX for future projects.

SOLAR BATTERIES AS AN ELEMENT OF DESIGN OF MODERN ENERGY EFFICIENT BUILDINGS

The concept of increasing the energy efficiency of buildings by placing solar panels on their facades is considered. This solution is evaluated from the point of view of modern urban design, and the possibilities of increasing the energy efficiency of lighting systems of modern buildings with the use of solar panels for their power supply are determined. The implementation of a combination of power from renewable energy sources and from the stationary grid in the absence of energy storage systems (batteries) allows to create systems capable of using almost all the electricity generated by an alternative energy source. In this case, energy losses to energy transportation will be minimal. The absence of such battery systems in the designs allows to install only the design-calculated number of solar panels without the need to make a significant backup of generating capacity for its accumulation or operation in adverse weather conditions. An additional advantage of abandoning energy storage - batteries, is a significant reduction in the use of environmentally hazardous substances such as lead, lithium and acidic or alkaline electrolytes, the production of which has a negative impact on the environment, and their disposal requires additional costs. In the work, it is proposed to use cassette facade systems with built-in solar panels, which allow architects to create a modern appearance of buildings and structures. It is shown that in the absence of solar tracking systems, the reduction in generation is 23% for the case of horizontal expansion of solar panels compared to the case of the optimal angle (for Kyiv 40o) and 40% for the case of vertical installation of solar panels. The analysis indicates the possibility of creating a modern energy-efficient design of building facades, both in the design of modern buildings and in the restoration of facades of existing buildings.