scMaui: a widely applicable deep learning framework for single-cell multiomics integration in the presence of batch effects and missing data.

The recent advances in high-throughput single-cell sequencing have created an urgent demand for computational models which can address the high complexity of single-cell multiomics data. Meticulous single-cell multiomics integration models are required to avoid biases towards a specific modality and overcome sparsity. Batch effects obfuscating biological signals must also be taken into account. Here, we introduce a new single-cell multiomics integration model, Single-cell Multiomics Autoencoder Integration (scMaui) based on variational product-of-experts autoencoders and adversarial learning. scMaui calculates a joint representation of multiple marginal distributions based on a product-of-experts approach which is especially effective for missing values in the modalities. Furthermore, it overcomes limitations seen in previous VAE-based integration methods with regard to batch effect correction and restricted applicable assays. It handles multiple batch effects independently accepting both discrete and continuous values, as well as provides varied reconstruction loss functions to cover all possible assays and preprocessing pipelines. We demonstrate that scMaui achieves superior performance in many tasks compared to other methods. Further downstream analyses also demonstrate its potential in identifying relations between assays and discovering hidden subpopulations.

An improved compound Poisson model for the number of motif hits in DNA sequences.

MOTIVATION: Transcription factors play a crucial role in gene regulation by binding to specific regulatory sequences. The sequence motifs recognized by a transcription factor can be described in terms of position frequency matrices. When scanning a sequence for matches to a position frequency matrix, one needs to determine a cut-off, which then in turn results in a certain number of hits. In this paper we describe how to compute the distribution of match scores and of the number of motif hits, which are the prerequisites to perform motif hit enrichment analysis. RESULTS: We put forward an improved compound Poisson model that supports general order-d Markov background models and which computes the number of motif-hits more accurately than earlier models. We compared the accuracy of the improved compound Poisson model with previously proposed models across a range of parameters and motifs, demonstrating the improvement. The importance of the order-d model is supported in a case study using CpG-island sequences. AVAILABILITY AND IMPLEMENTATION: The method is available as a Bioconductor package named 'motifcounter' https://bioconductor.org/packages/motifcounter. CONTACT: kopp@molgen.mpg.de. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Aging and "Age-Related" Diseases - What Is the Relation?

The study explores the intricate relationship between aging and the development of noncommunicable diseases [NCDs], focusing on whether these diseases are inevitable consequences of aging or primarily driven by lifestyle factors. By examining epidemiological data, particularly from hunter-gatherer societies, the study highlights that many NCDs prevalent in modern populations are rare in these societies, suggesting a significant influence of lifestyle choices. It delves into the mechanisms through which poor diet, smoking, and other lifestyle factors contribute to systemic physiological imbalances, characterized by oxidative stress, insulin resistance and hyperinsulinemia, and dysregulation of the sympathetic nervous system, the renin-angiotensin-aldosterone system, and the immune system. The interplay between this pattern and individual factors such as genetic susceptibility, biological variability, epigenetic changes and the microbiome is proposed to play a crucial role in the development of a range of age-related NCDs. Modified biomolecules such as oxysterols and advanced glycation end products also contribute to their development. Specific diseases such as benign prostatic hyperplasia, Parkinson's disease, glaucoma and osteoarthritis are analyzed to illustrate these mechanisms. The study concludes that while aging contributes to the risk of NCDs, lifestyle factors play a crucial role, offering potential avenues for prevention and intervention through healthier living practices. One possible approach could be to try to restore the physiological balance, e.g. through dietary measures [e.g. Mediterranean diet, Okinawan diet or Paleolithic diet] in conjunction with [a combination of] pharmacological interventions and other lifestyle changes.

Pathogenesis of (smoking-related) non-communicable diseases-Evidence for a common underlying pathophysiological pattern.

Non-communicable diseases, like diabetes, cardiovascular diseases, cancer, stroke, chronic obstructive pulmonary disease, osteoporosis, arthritis, Alzheimer's disease and other more are a leading cause of death in almost all countries. Lifestyle factors, especially poor diet and tobacco consumption, are considered to be the most important influencing factors in the development of these diseases. The Western diet has been shown to cause a significant distortion of normal physiology, characterized by dysregulation of the sympathetic nervous system, renin-angiotensin aldosterone system, and immune system, as well as disruption of physiological insulin and oxidant/antioxidant homeostasis, all of which play critical roles in the development of these diseases. This paper addresses the question of whether the development of smoking-related non-communicable diseases follows the same pathophysiological pattern. The evidence presented shows that exposure to cigarette smoke and/or nicotine causes the same complex dysregulation of physiology as described above, it further shows that the factors involved are strongly interrelated, and that all of these factors play a key role in the development of a broad spectrum of smoking-related diseases. Since not all smokers develop one or more of these diseases, it is proposed that this disruption of normal physiological balance represents a kind of pathogenetic "basic toolkit" for the potential development of a range of non-communicable diseases, and that the decision of whether and what disease will develop in an individual is determined by other, individual factors ("determinants"), such as the genome, epigenome, exposome, microbiome, and others. The common pathophysiological pattern underlying these diseases may provide an explanation for the often poorly understood links between non-communicable diseases and disease comorbidities. The proposed pathophysiological process offers new insights into the development of non-communicable diseases and may influence the direction of future research in both prevention and therapy.

Single-cell-resolved dynamics of chromatin architecture delineate cell and regulatory states in zebrafish embryos.

DNA accessibility of cis-regulatory elements (CREs) dictates transcriptional activity and drives cell differentiation during development. While many genes regulating embryonic development have been identified, the underlying CRE dynamics controlling their expression remain largely uncharacterized. To address this, we produced a multimodal resource and genomic regulatory map for the zebrafish community, which integrates single-cell combinatorial indexing assay for transposase-accessible chromatin with high-throughput sequencing (sci-ATAC-seq) with bulk histone PTMs and Hi-C data to achieve a genome-wide classification of the regulatory architecture determining transcriptional activity in the 24-h post-fertilization (hpf) embryo. We characterized the genome-wide chromatin architecture at bulk and single-cell resolution, applying sci-ATAC-seq on whole 24-hpf stage zebrafish embryos, generating accessibility profiles for ∼23,000 single nuclei. We developed a genome segmentation method, ScregSeg (single-cell regulatory landscape segmentation), for defining regulatory programs, and candidate CREs, specific to one or more cell types. We integrated the ScregSeg output with bulk measurements for histone post-translational modifications and 3D genome organization and identified new regulatory principles between chromatin modalities prevalent during zebrafish development. Sci-ATAC-seq profiling of npas4l/cloche mutant embryos identified novel cellular roles for this hematovascular transcriptional master regulator and suggests an intricate mechanism regulating its expression. Our work defines regulatory architecture and principles in the zebrafish embryo and establishes a resource of cell-type-specific genome-wide regulatory annotations and candidate CREs, providing a valuable open resource for genomics, developmental, molecular, and computational biology.

Multiomic atlas with functional stratification and developmental dynamics of zebrafish cis-regulatory elements.

Zebrafish, a popular organism for studying embryonic development and for modeling human diseases, has so far lacked a systematic functional annotation program akin to those in other animal models. To address this, we formed the international DANIO-CODE consortium and created a central repository to store and process zebrafish developmental functional genomic data. Our data coordination center ( https://danio-code.zfin.org ) combines a total of 1,802 sets of unpublished and re-analyzed published genomic data, which we used to improve existing annotations and show its utility in experimental design. We identified over 140,000 cis-regulatory elements throughout development, including classes with distinct features dependent on their activity in time and space. We delineated the distinct distance topology and chromatin features between regulatory elements active during zygotic genome activation and those active during organogenesis. Finally, we matched regulatory elements and epigenomic landscapes between zebrafish and mouse and predicted functional relationships between them beyond sequence similarity, thus extending the utility of zebrafish developmental genomics to mammals.

Predicting lethal courses in critically ill COVID-19 patients using a machine learning model trained on patients with non-COVID-19 viral pneumonia.

In a pandemic with a novel disease, disease-specific prognosis models are available only with a delay. To bridge the critical early phase, models built for similar diseases might be applied. To test the accuracy of such a knowledge transfer, we investigated how precise lethal courses in critically ill COVID-19 patients can be predicted by a model trained on critically ill non-COVID-19 viral pneumonia patients. We trained gradient boosted decision tree models on 718 (245 deceased) non-COVID-19 viral pneumonia patients to predict individual ICU mortality and applied it to 1054 (369 deceased) COVID-19 patients. Our model showed a significantly better predictive performance (AUROC 0.86 [95% CI 0.86-0.87]) than the clinical scores APACHE2 (0.63 [95% CI 0.61-0.65]), SAPS2 (0.72 [95% CI 0.71-0.74]) and SOFA (0.76 [95% CI 0.75-0.77]), the COVID-19-specific mortality prediction models of Zhou (0.76 [95% CI 0.73-0.78]) and Wang (laboratory: 0.62 [95% CI 0.59-0.65]; clinical: 0.56 [95% CI 0.55-0.58]) and the 4C COVID-19 Mortality score (0.71 [95% CI 0.70-0.72]). We conclude that lethal courses in critically ill COVID-19 patients can be predicted by a machine learning model trained on non-COVID-19 patients. Our results suggest that in a pandemic with a novel disease, prognosis models built for similar diseases can be applied, even when the diseases differ in time courses and in rates of critical and lethal courses.

Development of Obesity: The Driver and the Passenger.

Obesity has reached epidemic proportions and is one of the greatest challenges for public health in the twenty-first century. The macronutrient composition of diets, in particular the amount and ratio of carbohydrates, fat and protein, have received considerable attention in recent decades due to its potential relevance to the development of obesity and weight loss. The effects of various macronutrients on body weight regulation are still under debate. High-carbohydrate diets, and particularly high-fat diets, have been blamed for the increase in the prevalence of obesity. This paper shows that neither fat nor carbohydrates are fattening per se. Mixed diets with substantial amounts of fat and high-glycemic carbohydrates, like current WDs, are required to promote weight gain and obesity. High-glycemic carbohydrates are the active partner (the "driver"), which promotes fat storage through its insulinogenic effect, while fat is the passive partner (the "passenger") on the way to obesity. Elevated insulin levels (postprandial, but more importantly due to hypersecretion and hyperinsulinemia) promote fat storage and play a key role in obesogenesis and the obesity epidemic. Furthermore, mixed diets high in high-glycemic carbohydrates and fat promote fetal programming, with long-term adverse impacts on the offspring, including insulin hypersecretion, (childhood) obesity and metabolic diseases. Maternal obesity and high weight gain during pregnancy have also been linked to deleterious effects on fetal programming. As the global obesity epidemic increasingly affects women of reproductive age, a significant percentage of fetuses will experience fetal programming with a tendency towards obesity - a self-reinforcing process that further fuels the epidemic. A change in lifestyle and diet composition is needed to prevent or limit the development of obesity and related diseases.

Deep learning for genomics using Janggu.

In recent years, numerous applications have demonstrated the potential of deep learning for an improved understanding of biological processes. However, most deep learning tools developed so far are designed to address a specific question on a fixed dataset and/or by a fixed model architecture. Here we present Janggu, a python library facilitates deep learning for genomics applications, aiming to ease data acquisition and model evaluation. Among its key features are special dataset objects, which form a unified and flexible data acquisition and pre-processing framework for genomics data that enables streamlining of future research applications through reusable components. Through a numpy-like interface, these dataset objects are directly compatible with popular deep learning libraries, including keras or pytorch. Janggu offers the possibility to visualize predictions as genomic tracks or by exporting them to the bigWig format as well as utilities for keras-based models. We illustrate the functionality of Janggu on several deep learning genomics applications. First, we evaluate different model topologies for the task of predicting binding sites for the transcription factor JunD. Second, we demonstrate the framework on published models for predicting chromatin effects. Third, we show that promoter usage measured by CAGE can be predicted using DNase hypersensitivity, histone modifications and DNA sequence features. We improve the performance of these models due to a novel feature in Janggu that allows us to include high-order sequence features. We believe that Janggu will help to significantly reduce repetitive programming overhead for deep learning applications in genomics, and will enable computational biologists to rapidly assess biological hypotheses.

How Western Diet And Lifestyle Drive The Pandemic Of Obesity And Civilization Diseases.

Westernized populations are plagued by a plethora of chronic non-infectious degenerative diseases, termed as "civilization diseases", like obesity, diabetes, cardiovascular diseases, cancer, autoimmune diseases, Alzheimer's disease and many more, diseases which are rare or virtually absent in hunter-gatherers and other non-westernized populations. There is a growing awareness that the cause of this amazing discrepancy lies in the profound changes in diet and lifestyle during recent human history. This paper shows that the transition from Paleolithic nutrition to Western diets, along with lack of corresponding genetic adaptations, cause significant distortions of the fine-tuned metabolism that has evolved over millions of years of human evolution in adaptation to Paleolithic diets. With the increasing spread of Western diet and lifestyle worldwide, overweight and civilization diseases are also rapidly increasing in developing countries. It is suggested that the diet-related key changes in the developmental process include an increased production of reactive oxygen species and oxidative stress, development of hyperinsulinemia and insulin resistance, low-grade inflammation and an abnormal activation of the sympathetic nervous system and the renin-angiotensin system, all of which play pivotal roles in the development of diseases of civilization. In addition, diet-related epigenetic changes and fetal programming play an important role. The suggested pathomechanism is also able to explain the well-known but not completely understood close relationship between obesity and the wide range of comorbidities, like type 2 diabetes mellitus, cardiovascular disease, etc., as diseases of the same etiopathology. Changing our lifestyle in accordance with our genetic makeup, including diet and physical activity, may help prevent or limit the development of these diseases.

DNA Motif Match Statistics Without Poisson Approximation.

Transcription factors (TFs) play a crucial role in gene regulation by binding to specific regulatory sequences. The sequence motifs recognized by a TF can be described in terms of position frequency matrices. Searching for motif matches with a given position frequency matrix is achieved by employing a predefined score cutoff and subsequently counting the number of matches above this cutoff. In this article, we approximate the distribution of the number of motif matches based on a novel dynamic programming approach, which accounts for higher order sequence background (e.g., as is characteristic for CpG islands) and overlapping motif matches on both DNA strands. A comparison with our previously published compound Poisson approximation and a binomial approximation demonstrates that in particular for relaxed score thresholds, the dynamic programming approach yields more accurate results.

Diet-Induced Hyperinsulinemia as a Key Factor in the Etiology of Both Benign Prostatic Hyperplasia and Essential Hypertension?

Benign prostatic hyperplasia and hypertension are common age-related comorbidities. Although the etiology of benign prostatic hyperplasia (BPH) is still largely unresolved and poorly understood, a significant age-independent association was found between BPH and hypertension, indicating a common pathophysiological factor for both diseases. It has previously been suggested that the development of essential hypertension may be related to diet-induced hyperinsulinemia. This study follows the question, whether BPH may develop due to the same mechanism, thereby explaining the well-known comorbidity of these 2 disorders. The scientific evidence presented shows that BPH and hypertension share the same pathophysiological changes, with hyperinsulinemia as the driving force. It further shows that significant dietary changes during human history cause disruption of a finely tuned metabolic balance that has evolved over millions of years of evolution: high-insulinemic food, typical of current "Western" diets, has the potential to cause hyperinsulinemia and insulin resistance, as well as an abnormally increased activation of the sympathetic nervous system and the renin-angiotensin-aldosterone system, alterations that play a pivotal role in the pathogenesis of BPH and hypertension.

Esrrb Unlocks Silenced Enhancers for Reprogramming to Naive Pluripotency.

Transcription factor (TF)-mediated reprogramming to pluripotency is a slow and inefficient process, because most pluripotency TFs fail to access relevant target sites in a refractory chromatin environment. It is still unclear how TFs actually orchestrate the opening of repressive chromatin during the long latency period of reprogramming. Here, we show that the orphan nuclear receptor Esrrb plays a pioneering role in recruiting the core pluripotency factors Oct4, Sox2, and Nanog to inactive enhancers in closed chromatin during the reprogramming of epiblast stem cells. Esrrb binds to silenced enhancers containing stable nucleosomes and hypermethylated DNA, which are inaccessible to the core factors. Esrrb binding is accompanied by local loss of DNA methylation, LIF-dependent engagement of p300, and nucleosome displacement, leading to the recruitment of core factors within approximately 2 days. These results suggest that TFs can drive rapid remodeling of the local chromatin structure, highlighting the remarkable plasticity of stable epigenetic information.

Pathogenesis and etiology of essential hypertension: role of dietary carbohydrate.

The development of essential hypertension (EH) is proposed to be the result of a cascade of metabolic alterations, with high insulin levels/hyperinsulinemia and an abnormal reaction to the vasodilatory effect of insulin as the initiating factors. It is well established that insulin causes vasodilatation of peripheral resistance vessels. In normal subjects, this insulin-induced vasodilatation and decrease of the peripheral vascular resistance (PVR) is compensated by an SNS-mediated re-vasoconstriction in order to avoid hypotension, with the net effect of a slight decrease in blood pressure and no significant effect on peripheral vascular resistance. In contrast, in genetically predisposed subjects, prone to the development of essential hypertension, the insulin-induced vasodilatation is compensated by an increased heart rate and cardiac output (to avoid hypotension), mediated by an abnormal sympathetic overactivity, (characterised by high norepinephrine spillover rates and (frequently) a hyperdynamic circulation), while the PVR remains low during the early phase of developing EH. During the course of chronic hypertension, the SNS-overactivity leads to progressive trophic alterations of vessel walls, and structural and functional vascular remodeling, with narrowing of arterial resistance vessels and an increasing PVR. Vascular remodeling and lumen narrowing not only affect peripheral resistance vessels, but also kidney vessels. Narrowing and decreased distensibility of preglomerular kidney vessels lead to chronic activation of the Renin-Angiotensin-Aldosterone-System, with reinforcement and fixation of hypertension. High-glycemic index nutrition is suggested to play a key role in the etiology of hypertension: The chronic stimulus of pancreatic beta-cells due to high-glycemic index nutrition may cause cell hypertrophy and dysfunction, resulting in postprandial hyperinsulinemia, and -- in susceptible subjects -- the development of EH. Since significant evidence suggests that hyperinsulinemia also represents a key factor for the development of obesity, insulin resistance and the metabolic syndrome, the well-known common association of EH and these metabolic alterations becomes quite understandable.

Role of high-insulinogenic nutrition in the etiology of gestational diabetes mellitus.

A transient physiologic insulin resistance and hyperinsulinemia are characteristic of normal pregnancy. This insulin action has evolved during a period of human evolution that was characterized by a very low-carbohydrate nutrition. The development of gestational diabetes mellitus (GDM) is proposed to result from a collision of this evolutionary inheritance with our "modern" nutrition: The "Western" high-insulinogenic nutrition increases the postprandial demand for insulin significantly during the insulin resistant state of late pregnancy. In women with beta-cells that are not capable of maintaining the high insulin production, GDM develops. A restriction of high-insulinogenic carbohydrate may help to prevent the development of GDM.

High-insulinogenic nutrition--an etiologic factor for obesity and the metabolic syndrome?

This report postulates a critical role for the quantity and quality of dietary carbohydrate in the pathogenesis of obesity and the metabolic syndrome. Significant changes in human nutrition have occurred during the last 10,000 years, culminating in the current high-glycemic/high-insulinogenic nutrition. A high insulinogenic nutrition represents a chronic stimulus to the beta cells that may induce an adaptive hypertrophy and a progressive dysregulation of the cells, resulting in postprandial hyperinsulinemia, especially in genetically predisposed subjects. Significant evidence suggests that postprandial hyperinsulinemia promotes weight gain and the development of insulin resistance/metabolic syndrome. The hypothesis is able to explain the current epidemic of obesity and the metabolic syndrome in most industrialised countries, as well as some of the genetics of obesity, including the extreme high incidence of obesity and the metabolic syndrome in certain ethnic groups.

Nutrition, evolution and thyroid hormone levels - a link to iodine deficiency disorders?

An increased iodine requirement as a result of significant changes in human nutrition rather than a decreased environmental iodine supply is suggested to represent the main cause of the iodine deficiency disorders (IDD). The pathomechanism proposed is based on the fact that serum concentrations of thyroid hormones, especially of trijodothyronine (T3), are dependent on the amount of dietary carbohydrate. High-carbohydrate diets are associated with significantly higher serum T3 concentrations, compared with very low-carbohydrate diets. While our Paleolithic ancestors subsisted on a very low carbohydrate/high protein diet, the agricultural revolution about 10,000 years ago brought about a significant increase in dietary carbohydrate. These nutritional changes have increased T3 levels significantly. Higher T3 levels are associated with an enhanced T3 production and an increased iodine requirement. The higher iodine requirement exceeds the availability of iodine from environmental sources in many regions of the world, resulting in the development of IDD.

The atherogenic potential of dietary carbohydrate.

OBJECTIVE: To investigate the role of dietary carbohydrate in atherogenesis. METHOD: Search of the literature for relevant papers concerning the relationship of insulin/hyperinsulinemia and carbohydrate on the one hand, and the renin-angiotensin system, the sympathetic nervous system, growth factors, i.e. platelet-derived growth factor and insulin-like growth factor-I, C-reactive protein, and dyslipemia, on the other hand, factors well known to be involved in the atherogenic process, as well as for epidemiologic studies investigating the relationship between high-carbohydrate diets and the development of cardiovascular disease. RESULTS: High-carbohydrate nutrition is shown to have the ability to induce vascular inflammation and plaque formation through an insulin-mediated activation of the RAS, growth factors, cytokines, the SNS, and C-reactive protein and to cause an atherogenic lipid profile in normal humans. Epidemiologic studies as well as studies in experimental animals corroborate an important role of dietary carbohydrate in atherogenesis. CONCLUSION: High-carbohydrate diets, particularly in the form of high-glycemic index carbohydrate, have the ability to directly induce atherosclerosis. Based on anthropologic facts, the reason for these dietary-induced, insulin-mediated, atherogenic metabolic perturbations are suggested to be an insufficient adaptation to starch and sugars during human evolution. Restriction of insulinogenic food (starch and sugars) may help to prevent the development of atherosclerosis, one of the most common and costliest human diseases.

'Arzteführerschule Alt-Rehse'--the main educational centre for promotion of Nazi medicine ideology in Germany during 1935 - 1943.

In Nazi Germany 1933-1945 the medical profession played an important role, not only within the health sector but also for ideological purposes. Some physicians committed criminal acts during the war, for example in concentration camps, and where later put on trial at the Nuremberg Trials in 1946-1947, some of them also sentenced to death. What is little known outside Germany is the existence of a specialised political leader school for physicians, dentists, midwives, pharmacists, and health officials at Alt-Rehse, a small village north of Berlin in the province of Mecklenburg, Northern Germany. Supported by historical documents it can be shown that the ideological training at Alt-Rehse was structured, politically effective, and well appreciated by the many attending students visiting the school. There is reason to believe that former Alt-Rehse students took part in Nazi criminal acts of war, but this is hard to prove as all documents were captured by the Russian Army at the end of the Second World War and have until now not been traced. Documents have shown that also many foreign visitors came to Alt-Rehse, from the UK, Sweden, Japan, and other countries during the pre-war years. The lessons from Alt-Rehse include how a totalitarian ideology can shape the professional role of health workers.