Benchmarking atlas-level data integration in single-cell genomics.

Single-cell atlases often include samples that span locations, laboratories and conditions, leading to complex, nested batch effects in data. Thus, joint analysis of atlas datasets requires reliable data integration. To guide integration method choice, we benchmarked 68 method and preprocessing combinations on 85 batches of gene expression, chromatin accessibility and simulation data from 23 publications, altogether representing >1.2 million cells distributed in 13 atlas-level integration tasks. We evaluated methods according to scalability, usability and their ability to remove batch effects while retaining biological variation using 14 evaluation metrics. We show that highly variable gene selection improves the performance of data integration methods, whereas scaling pushes methods to prioritize batch removal over conservation of biological variation. Overall, scANVI, Scanorama, scVI and scGen perform well, particularly on complex integration tasks, while single-cell ATAC-sequencing integration performance is strongly affected by choice of feature space. Our freely available Python module and benchmarking pipeline can identify optimal data integration methods for new data, benchmark new methods and improve method development.

Artificial Intelligence in Orthodontics: Critical Review.

With increasing digitalization in orthodontics, certain orthodontic manufacturing processes such as the fabrication of indirect bonding trays, aligner production, or wire bending can be automated. However, orthodontic treatment planning and evaluation remains a specialist's task and responsibility. As the prediction of growth in orthodontic patients and response to orthodontic treatment is inherently complex and individual, orthodontists make use of features gathered from longitudinal, multimodal, and standardized orthodontic data sets. Currently, these data sets are used by the orthodontist to make informed, rule-based treatment decisions. In research, artificial intelligence (AI) has been successfully applied to assist orthodontists with the extraction of relevant data from such data sets. Here, AI has been applied for the analysis of clinical imagery, such as automated landmark detection in lateral cephalograms but also for evaluation of intraoral scans or photographic data. Furthermore, AI is applied to help orthodontists with decision support for treatment decisions such as the need for orthognathic surgery or for orthodontic tooth extractions. One major challenge in current AI research in orthodontics is the limited generalizability, as most studies use unicentric data with high risks of bias. Moreover, comparing AI across different studies and tasks is virtually impossible as both outcomes and outcome metrics vary widely, and underlying data sets are not standardized. Notably, only few AI applications in orthodontics have reached full clinical maturity and regulatory approval, and researchers in the field are tasked with tackling real-world evaluation and implementation of AI into the orthodontic workflow.

Publicly Available Dental Image Datasets for Artificial Intelligence.

The development of artificial intelligence (AI) in dentistry requires large and well-annotated datasets. However, the availability of public dental imaging datasets remains unclear. This study aimed to provide a comprehensive overview of all publicly available dental imaging datasets to address this gap and support AI development. This observational study searched all publicly available dataset resources (academic databases, preprints, and AI challenges), focusing on datasets/articles from 2020 to 2023, with PubMed searches extending back to 2011. We comprehensively searched for dental AI datasets containing images (intraoral photos, scans, radiographs, etc.) using relevant keywords. We included datasets of >50 images obtained from publicly available sources. We extracted dataset characteristics, patient demographics, country of origin, dataset size, ethical clearance, image details, FAIRness metrics, and metadata completeness. We screened 131,028 records and extracted 16 unique dental imaging datasets. The datasets were obtained from Kaggle (18.8%), GitHub, Google, Mendeley, PubMed, Zenodo (each 12.5%), Grand-Challenge, OSF, and arXiv (each 6.25%). The primary focus was tooth segmentation (62.5%) and labeling (56.2%). Panoramic radiography was the most common imaging modality (58.8%). Of the 13 countries, China contributed the most images (2,413). Of the datasets, 75% contained annotations, whereas the methods used to establish labels were often unclear and inconsistent. Only 31.2% of the datasets reported ethical approval, and 56.25% did not specify a license. Most data were obtained from dental clinics (50%). Intraoral radiographs had the highest findability score in the FAIR assessment, whereas cone-beam computed tomography datasets scored the lowest in all categories. These findings revealed a scarcity of publicly available imaging dental data and inconsistent metadata reporting. To promote the development of robust, equitable, and generalizable AI tools for dental diagnostics, treatment, and research, efforts are needed to address data scarcity, increase diversity, mandate metadata completeness, and ensure FAIRness in AI dental imaging research.

[Validation of the German version of the Singing Voice Handicap Index]. / Validierung des Singing Voice Handicap Index in der deutschen Fassung.

The Singing Voice Handicap Index (SVHI) was developed in the United States for the self-assessment of patients with singing problems. It has been translated into German and its reliability and validity have been assessed. In total, 54 (35 female, 19 male) dysphonic singers and 130 (74 female, 56 male) non-dysphonic professional singers were included in the study. Reliability rested on high test-retest reliability (r = 0.960, p ≤ 0.001, Pearson correlation) and a Cronbach's α of 0.975. A principal component analysis using the Varimax method and the results of the screeplot suggest the SVHI scored as a single scale. Validity rested on a highly significant correlation between the severity of the self-rated voice impairment by the patient and the total SVHI score. Dysphonic singers have significantly higher SVHI scores than healthy singers. The SVHI is thus suited to implementation as a diagnostic tool in German-speaking countries.

Multi-modal deep learning for automated assembly of periapical radiographs.

OBJECTIVES: Periapical radiographs are oftentimes taken in series to display all teeth present in the oral cavity. Our aim was to automatically assemble such a series of periapical radiographs into an anatomically correct status using a multi-modal deep learning model. METHODS: 4,707 periapical images from 387 patients (on average, 12 images per patient) were used. Radiographs were labeled according to their field of view and the dataset split into a training, validation, and test set, stratified by patient. In addition to the radiograph the timestamp of image generation was extracted and abstracted as follows: A matrix, containing the normalized timestamps of all images of a patient was constructed, representing the order in which images were taken, providing temporal context information to the deep learning model. Using the image data together with the time sequence data a multi-modal deep learning model consisting of two residual convolutional neural networks (ResNet-152 for image data, ResNet-50 for time data) was trained. Additionally, two uni-modal models were trained on image data and time data, respectively. A custom scoring technique was used to measure model performance. RESULTS: Multi-modal deep learning outperformed both uni-modal image-based learning (p<0.001) and time-based learning (p<0.05). The multi-modal deep learning model predicted tooth labels with an F1-score, sensitivity and precision of 0.79, respectively, and an accuracy of 0.99. 37 out of 77 patient datasets were fully correctly assembled by multi-modal learning; in the remaining ones, usually only one image was incorrectly labeled. CONCLUSIONS: Multi-modal modeling allowed automated assembly of periapical radiographs and outperformed both uni-modal models. Dental machine learning models can benefit from additional data modalities. CLINICAL SIGNIFICANCE: Like humans, deep learning models may profit from multiple data sources for decision-making. We demonstrate how multi-modal learning can assist assembling periapical radiographs into an anatomically correct status. Multi-modal learning should be considered for more complex tasks, as clinically a wealth of data is usually available and could be leveraged.

Microbial contamination on ambulance surfaces: a systematic literature review.

Healthcare-associated infections (HAIs) are infections that patients acquire while receiving medical treatment in a healthcare facility. During ambulatory transport, the patient may be exposed to pathogens transmitted from emergency medical service (EMS) personnel or EMS surfaces.The aim of this study was to determine whether organisms commonly associated with HAIs have been detected on surfaces in the patient-care compartment of ambulances. Five electronic databases - PubMed, Scopus, Web of Science, Embase and Google Scholar were used to search for articles using inclusion and exclusion criteria following the PRISMA checklist. Inclusion criteria consisted of articles published in English, between 2009 and 2020, had positive samples collected from the patient-care compartment of a ground ambulance, and reported sample collection methods of either swab sampling and/or Replicate Organism Detection and Counting (RODAC) contact plates. Studies not meeting these criteria were excluded from this review. From a total of 1376 articles identified, 16 were included in the review. Organisms associated with HAIs were commonly detected in the patient-care compartment of ambulances across a variety of different surfaces, including blood pressure cuffs, oxygen apparatuses, and areas of patient stretchers. A high prevalence of pathogenic bacteria in ambulances suggests that standard protocols related to cleaning compliance may not be effective. The primary recommendation is that designated subject matter experts in infection prevention should be incorporated as liaisons in the pre-hospital setting, acting as a link between the pre-hospital (e.g., ambulance transport) and hospital environments.

Original findings associated with two cases of bovine papular stomatitis.

Bovine papular stomatitis virus was isolated from two calves in an animal house with biosafety level 3 confinement. The hypotheses on the origin of the infection, the interesting features of the partial amino acid sequences of the major envelope viral protein, and the importance of diagnostic tools available for animal diseases that are not listed by the World Organization for Animal Health (OIE) are discussed.

[Evaluation of a renal biopsy: what information is important for nephrologists?]. / Beurteilung der Nierenbiopsie: Was der Nephrologe vom (Nephro-)Pathologen wissen will.

A kidney biopsy is an important and frequently used diagnostic tool in routine nephrology. In order to obtain relevant clinical information from a renal biopsy close cooperation between clinicians and pathologists is mandatory. The better the information obtained from nephrologists and the better the understanding by nephrologists and the quality of the kidney biopsy, the more rewarding is the information from pathologists. The following paper will discuss some practical aspects regarding the interaction between nephrology and pathology which may not be known or poorly handled and may thus cause misunderstanding. In order to facilitate interaction between clinicians and pathologists some guidelines concerning the procedure and work-up of routine kidney biopsies have been established and will be discussed in detail.

[Kidney transplant pathology]. / Nierentransplantatpathologie.

Biopsy of the transplanted kidney plays an important role in the care and treatment of patients after kidney transplantation. Today the renal biopsy is a standard procedure which is performed early after renal transplantation in the case of a primary non-functioning graft or a significant rise in serum creatinine. On the other hand, a kidney biopsy is performed if an acute or creeping rise in serum creatinine or acute onset of proteinuria or erythrocyturia is observed during follow-up. Furthermore, zero biopsies or intraoperative biopsies of the graft are important in order to obtain information about the initial quality of the graft. This is particularly important in view of the shortage of donor organs and the resulting necessity to accept increasingly marginal organs, such as for example in the ESP program. In addition, an increasing number of transplant centres perform protocol biopsies, i.e. biopsies that are not based on clinical indication, but are performed at a certain time point after transplantation to detect subclinical rejections as well as histological alterations pointing to chronic allograft damage. Additionally, there is much scientific interest in protocol biopsies.

scCODA is a Bayesian model for compositional single-cell data analysis.

Compositional changes of cell types are main drivers of biological processes. Their detection through single-cell experiments is difficult due to the compositionality of the data and low sample sizes. We introduce scCODA ( https://github.com/theislab/scCODA ), a Bayesian model addressing these issues enabling the study of complex cell type effects in disease, and other stimuli. scCODA demonstrated excellent detection performance, while reliably controlling for false discoveries, and identified experimentally verified cell type changes that were missed in original analyses.

Tiny golden angle stack-of-stars (tygaSoS) free-breathing functional lung imaging.

PURPOSE: MRI of the lung parenchyma is still challenging due to cardiac and respiratory motion, and the low proton density and short T2*. Clinical feasible MRI methods for functional lung assessment are of great interest. It was the objective of this study to evaluate the potential of combining the ultra-short echo-time stack-of-stars approach with tiny golden angle (tyGASoS) profile ordering for self-gated free-breathing lung imaging. METHODS: Free-breathing tyGASoS data were acquired in 10 healthy volunteers (3 smoker (S), 7 non-smoker (NS)). Images in different respiratory phases were reconstructed applying an image-based self-gating technique. Resulting image quality and sharpness, and parenchyma visibility were qualitatively scored by three blinded independent reader, and the signal-to-noise ratio (SNR), proton fraction (fP) and fractional ventilation (FV) quantified. RESULT: The imaging protocol was well tolerated by all volunteers. Image quality was sufficient for subsequent quantitative analysis in all cases with good to excellent inter-reader reliability. Between expiration (EX) and inspiration (IN) significant differences (p < 0.001) were observed in SNR (EX: 3.73 ± 0.89, IN: 3.14 ± 0.74) and fP (EX: 0.27 ± 0.09, IN: 0.25 ± 0.08). A significant (p < 0.05) higher fP (EX/IN: 0.22 ± 0.07/0.21 ± 0.07 (NS), 0.33 ± 0.07/0.30 ± 0.06 (S)) was observed in the smoker group. No significant FV differences resulted between S and NS. CONCLUSION: The study proves the feasibility of free-breathing tyGASoS for multiphase lung imaging. Changes in fP may indicate an initial response in the smoker group and as such proves the sensitivity of the proposed technique. A major limitation in FV quantification rises from the large inter-subject variability of breathing patterns and amplitudes, requiring further consideration.

Immunohistological studies on neoplasms of female and male Onchocerca volvulus: filarial origin and absence of Wolbachia from tumor cells.

Up to 5% of untreated female Onchocerca volvulus filariae develop potentially fatal pleomorphic neoplasms, whose incidence is increased following ivermectin treatment. We studied the occurrence of 8 filarial proteins and of Wolbachia endobacteria in the tumor cells. Onchocercomas from patients, untreated and treated with antibiotics and anthelminthics, were examined by immunohistology. Neoplasms were diagnosed in 112 of 3587 female and in 2 of 1570 male O. volvulus. The following proteins and other compounds of O. volvulus were expressed in the cells of the neoplasms: glutathione S-transferase 1, lysosomal aspartic protease, cAMP-dependent protein kinase, alpha-enolase, aspartate aminotransferase, ankyrin E1, tropomyosin, heat shock protein 60, transforming growth factor-beta, and prostaglandin E(2). These findings prove the filarial origin of the neoplasms and confirm the pleomorphism of the tumor cells. Signs indicating malignancy of the neoplasms are described. Wolbachia were observed in the hypodermis, oocytes, and embryos of tumor-harbouring filariae using antibodies against Wolbachia surface protein, Wolbachia HtrA-type serine protease, and Wolbachia aspartate aminotransferase. In contrast, Wolbachia were not found in the cells of the neoplasms. Further, neoplasm-containing worms were not observed after more than 10 months after the start of sufficient treatment with doxycycline or doxycycline plus ivermectin.

The Region of Difference Four is a Robust Genetic Marker for Subtyping Mycobacterium caprae Isolates and is Linked to Spatial Distribution of Three Subtypes.

Alpine Mycobacterium caprae isolates found in cattle and red deer display at least three genetic variations in the region of difference four (RD4) that can be used for further differentiation of the isolates into the subtypes 'Allgäu', 'Karwendel' and 'Lechtal'. Each genomic subtype is thereby characterized by a specific nucleotide deletion pattern in the 12.7-kb RD4 region. Even though M. caprae infections are frequently documented in cattle and red deer, little is known about the transmission routes. Hence, robust markers for M. caprae subtyping are needed to gain insight into the molecular epidemiology. For this reason, a rapid and robust multiplex PCR was developed for the simultaneous detection of three M. caprae RD4 subtypes and was used to subtype a total number of 241 M. caprae isolates from animals (145 cattle, 95 red deer and one fox) from Bavaria and Austria. All three subtypes occur spatially distributed and are found in cattle and in red deer suggesting transmission between the two species. As subtypes are genetically stable in both species it is hypothesized that the described genetic variations developed within the host due to 'within-host replication'. The results of this study recommend the genomic RD4 region as a reliable diagnostic marker for M. caprae subtype differentiation.

Initiation of aerial mycelium formation in Streptomyces.

In the past two years, the isolation of extracellular factors involved in the initiation of aerial mycelium formation, the identification of metabolic defects in certain developmental mutants, and the characterisation of three further bld genes and several gamma-butyrolactone receptor genes have led to new ideas about the mechanisms that initiate aerial mycelium formation in Streptomyces. The emerging picture suggests the integration of numerous signals from both inside and outside the cell.

Influenza A Viruses Detected in Swine in Southern Germany after the H1N1 Pandemic in 2009.

Infections with influenza A viruses (IAV) are highly prevalent in swine populations, and stable cocirculation of at least three lineages has been well documented in European swine - till 2009. However, since the emergence of the human pandemic pdmH1N1 virus in 2009, which has been (re)introduced into individual swine herds worldwide, the situation has been changing. These variations in the respective IAV pools within pig populations are of major interest, and the zoonotic potential of putative emerging viruses needs to be evaluated. As data on recent IAV in swine from southern Germany were relatively sparse, the purpose of this study was to determine the major IAV subtypes actually present in this region. To this aim, from 2010 to 2013, 1417 nasal swabs or lung tissue samples from pigs with respiratory disease were screened for IAV genomes. Overall, in 130 holdings IAV genomes were detected by real-time RT-PCR targeting the matrix protein gene. For further analyses, several PCR protocols were adapted to quickly subtype between H1, pdmH1, H3, N1 and N2 sequences. Taken together, cocirculation of the three stable European lineages of IAV was confirmed for Bavaria. H1N1 sequences were identified in 59, whereas H1N2 genomes were only diagnosed in 14, and H3N2 in 9 of the holdings analysed. However, pdmH1 in combination with N1 was detected in 2010, 2012 and 2013 confirming a presence, albeit in low prevalence, likewise pdmH1N2 reassortant viruses. Interestingly, individual cases of coinfections with more than one subtype were diagnosed. Partial genome sequences were determined and phylogenetic analyses performed. Clearly other than in the human population classically circulating IAV have not been displaced by pdmH1N1 in Bavarian swine. However, some interesting viruses were detected. Further surveillance of these viruses in the Bavarian pig population will be of major importance, to monitor future developments.

Monosaccharide transporters in plants: structure, function and physiology.

Monosaccharide transport across the plant plasma membrane plays an important role both in lower and higher plants. Algae can switch between phototrophic and heterotrophic growth and utilize organic compounds, such as monosaccharides as additional or sole carbon sources. Higher plants represent complex mosaics of phototrophic and heterotrophic cells and tissues and depend on the activity of numerous transporters for the correct partitioning of assimilated carbon between their different organs. The cloning of monosaccharide transporter genes and cDNAs identified closely related integral membrane proteins with 12 transmembrane helices exhibiting significant homology to monosaccharide transporters from yeast, bacteria and mammals. Structural analyses performed with several members of this transporter superfamily identified protein domains or even specific amino acid residues putatively involved in substrate binding and specificity. Expression of plant monosaccharide transporter cDNAs in yeast cells and frog oocytes allowed the characterization of substrate specificities and kinetic parameters. Immunohistochemical studies, in situ hybridization analyses and studies performed with transgenic plants expressing reporter genes under the control of promoters from specific monosaccharide transporter genes allowed the localization of the transport proteins or revealed the sites of gene expression. Higher plants possess large families of monosaccharide transporter genes and each of the encoded proteins seems to have a specific function often confined to a limited number of cells and regulated both developmentally and by environmental stimuli.

The photosystem I-like P840-reaction center of green S-bacteria is a homodimer.

An operon encoding the P840 reaction center of Chlorobium limicola f.sp.thiosulfatophilum has been cloned and sequenced. It contains two structural genes coding for proteins of 730 and 232 amino acids. The first protein resembles the large subunits of the Photosystem I (PS I) reaction center. Putative binding elements for the primary donor, P840 in Chlorobium and P700 in PS I and for the acceptors A(o), A(1) and FeS-center X are conserved. The second protein is related to the PS I subunit carrying the FeS-centers A and B. Since all our efforts to find a gene for a second, large subunit failed, the P840 reaction center probably is homodimeric.

RNA polymerase-DNA interactions in Streptomyces. In vitro studies of a S. lividans plasmid promoter with S. coelicolor RNA polymerase.

DNA fragments of the Streptomyces lividans plasmid pIJ101 have been tested for their ability to bind Streptomyces coelicolor RNA polymerase in vitro or to promote transcription in Streptomyces in vivo. One DNA fragment which does both was shown to encode a transcript which was expressed at low cell-density in cultures of pIJ101-containing cells. The transcript start was located on the DNA sequence of the fragment by nucleotide-primed RNA polymerase binding experiments and by S1 nuclease mapping. The pattern of DNase I protection, the sites of enhanced DNase I cleavage and the DNA sequence of the fragment suggest that the RNA polymerase holoenzyme form, which recognizes this promoter, is similar in its interaction with DNA to the major RNA polymerase of Escherichia coli. Regions showing 3/6 nucleotide homology with each of the -35 and -10 regions of the consensus sequence of E. coli promoters are present in the same positions relative to the transcript start. Symmetrical sequences which may be involved in the regulation of expression of the promoter and a potential polypeptide coding sequence can be identified.

Expression of a plant cell wall invertase in roots of Arabidopsis leads to early flowering and an increase in whole plant biomass.

In order to enhance sink strength, we expressed a heterologous plant cell wall invertase (CrCIN1) under the control of a root-specific promoter (ppyk10) in Arabidopsis thaliana. Slightly elevated apoplastic invertase activity resulted in apparent phenotypic changes. Transgenic plants developed more secondary roots and subsequently, possibly because of a higher capacity to acquire nutrients, a higher shoot and whole plant biomass. Furthermore, an early flowering phenotype was detected. The data presented here demonstrate that it is possible to modulate carbohydrate metabolism by ectopic expression of cell wall invertases and thereby influence sink organ size and whole plant development.

Some physico-chemical parameters that influence proteinase K resistance and the infectivity of PrPSc after high pressure treatment.

Crude brain homogenates of terminally diseased hamsters infected with the 263 K strain of scrapie (PrP Sc) were heated and/or pressurized at 800 MPa at 60 degrees C for different times (a few seconds or 5, 30, 120 min) in phosphate-buffered saline (PBS) of different pH and concentration. Prion proteins were analyzed on immunoblots for their proteinase K (PK) resistance, and in hamster bioassays for their infectivity. Samples pressurized under initially neutral conditions and containing native PrP Sc were negative on immunoblots after PK treatment, and a 6-7 log reduction of infectious units per gram was found when the samples were pressurized in PBS of pH 7.4 for 2 h. A pressure-induced change in the protein conformation of native PrP Sc may lead to less PK resistant and less infectious prions. However, opposite results were obtained after pressurizing native infectious prions at slightly acidic pH and in PBS of higher concentration. In this case an extensive fraction of native PrP Sc remained PK resistant after pressure treatment, indicating a protective effect possibly due to induced aggregation of prion proteins in such buffers.