Histological classification of canine and feline lymphoma using a modular approach based on deep learning and advanced image processing.

Histopathological examination of tissue samples is essential for identifying tumor malignancy and the diagnosis of different types of tumor. In the case of lymphoma classification, nuclear size of the neoplastic lymphocytes is one of the key features to differentiate the different subtypes. Based on the combination of artificial intelligence and advanced image processing, we provide a workflow for the classification of lymphoma with regards to their nuclear size (small, intermediate, and large). As the baseline for our workflow testing, we use a Unet++ model trained on histological images of canine lymphoma with individually labeled nuclei. As an alternative to the Unet++, we also used a publicly available pre-trained and unmodified instance segmentation model called Stardist to demonstrate that our modular classification workflow can be combined with different types of segmentation models if they can provide proper nuclei segmentation. Subsequent to nuclear segmentation, we optimize algorithmic parameters for accurate classification of nuclear size using a newly derived reference size and final image classification based on a pathologists-derived ground truth. Our image classification module achieves a classification accuracy of up to 92% on canine lymphoma data. Compared to the accuracy ranging from 66.67 to 84% achieved using measurements provided by three individual pathologists, our algorithm provides a higher accuracy level and reproducible results. Our workflow also demonstrates a high transferability to feline lymphoma, as shown by its accuracy of up to 84.21%, even though our workflow was not optimized for feline lymphoma images. By determining the nuclear size distribution in tumor areas, our workflow can assist pathologists in subtyping lymphoma based on the nuclei size and potentially improve reproducibility. Our proposed approach is modular and comprehensible, thus allowing adaptation for specific tasks and increasing the users' trust in computer-assisted image classification.

The Class IIA Histone Deacetylase (HDAC) Inhibitor TMP269 Downregulates Ribosomal Proteins and Has Anti-Proliferative and Pro-Apoptotic Effects on AML Cells.

Acute myeloid leukemia (AML) is a hematopoietic malignancy characterized by altered myeloid progenitor cell proliferation and differentiation. As in many other cancers, epigenetic transcriptional repressors such as histone deacetylases (HDACs) are dysregulated in AML. Here, we investigated (1) HDAC gene expression in AML patients and in different AML cell lines and (2) the effect of treating AML cells with the specific class IIA HDAC inhibitor TMP269, by applying proteomic and comparative bioinformatic analyses. We also analyzed cell proliferation, apoptosis, and the cell-killing capacities of TMP269 in combination with venetoclax compared to azacitidine plus venetoclax, by flow cytometry. Our results demonstrate significantly overexpressed class I and class II HDAC genes in AML patients, a phenotype which is conserved in AML cell lines. In AML MOLM-13 cells, TMP269 treatment downregulated a set of ribosomal proteins which are overexpressed in AML patients at the transcriptional level. TMP269 showed anti-proliferative effects and induced additive apoptotic effects in combination with venetoclax. We conclude that TMP269 exerts anti-leukemic activity when combined with venetoclax and has potential as a therapeutic drug in AML.

ImmunoDataAnalyzer: a bioinformatics pipeline for processing barcoded and UMI tagged immunological NGS data.

BACKGROUND: Next-generation sequencing (NGS) is nowadays the most used high-throughput technology for DNA sequencing. Among others NGS enables the in-depth analysis of immune repertoires. Research in the field of T cell receptor (TCR) and immunoglobulin (IG) repertoires aids in understanding immunological diseases. A main objective is the analysis of the V(D)J recombination defining the structure and specificity of the immune repertoire. Accurate processing, evaluation and visualization of immune repertoire NGS data is important for better understanding immune responses and immunological behavior. RESULTS: ImmunoDataAnalyzer (IMDA) is a pipeline we have developed for automatizing the analysis of immunological NGS data. IMDA unites the functionality from carefully selected immune repertoire analysis software tools and covers the whole spectrum from initial quality control up to the comparison of multiple immune repertoires. It provides methods for automated pre-processing of barcoded and UMI tagged immune repertoire NGS data, facilitates the assembly of clonotypes and calculates key figures for describing the immune repertoire. These include commonly used clonality and diversity measures, as well as indicators for V(D)J gene segment usage and between sample similarity. IMDA reports all relevant information in a compact summary containing visualizations, calculations, and sample details, all of which serve for a more detailed overview. IMDA further generates an output file including key figures for all samples, designed to serve as input for machine learning frameworks to find models for differentiating between specific traits of samples. CONCLUSIONS: IMDA constructs TCR and IG repertoire data from raw NGS reads and facilitates descriptive data analysis and comparison of immune repertoires. The IMDA workflow focus on quality control and ease of use for non-computer scientists. The provided output directly facilitates the interpretation of input data and includes information about clonality, diversity, clonotype overlap as well as similarity, and V(D)J gene segment usage. IMDA further supports the detection of sample swaps and cross-sample contamination that potentially occurred during sample preparation. In summary, IMDA reduces the effort usually required for immune repertoire data analysis by providing an automated workflow for processing raw NGS data into immune repertoires and subsequent analysis. The implementation is open-source and available on https://bioinformatics.fh-hagenberg.at/immunoanalyzer/ .

Prospective Tracking of Donor-Reactive T-Cell Clones in the Circulation and Rejecting Human Kidney Allografts.

Background: Antigen recognition of allo-peptides and HLA molecules leads to the activation of donor-reactive T-cells following transplantation, potentially causing T-cell-mediated rejection (TCMR). Sequencing of the T-cell receptor (TCR) repertoire can be used to track the donor-reactive repertoire in blood and tissue of patients after kidney transplantation. Methods/Design: In this prospective cohort study, 117 non-sensitized kidney transplant recipients with anti-CD25 induction were included. Peripheral mononuclear cells (PBMCs) were sampled pre-transplant and at the time of protocol or indication biopsies together with graft tissue. Next-generation sequencing (NGS) of the CDR3 region of the TCRbeta chain was performed after donor stimulation in mixed lymphocyte reactions to define the donor-reactive TCR repertoire. Blood and tissue of six patients experiencing a TCMR and six patients without rejection on protocol biopsies were interrogated for these TCRs. To elucidate common features of T-cell clonotypes, a network analysis of the TCR repertoires was performed. Results: After transplantation, the frequency of circulating donor-reactive CD4 T-cells increased significantly from 0.86 ± 0.40% to 2.06 ± 0.40% of all CD4 cells (p < 0.001, mean dif.: -1.197, CI: -1.802, -0.593). The number of circulating donor-reactive CD4 clonotypes increased from 0.72 ± 0.33% to 1.89 ± 0.33% (p < 0.001, mean dif.: -1.168, CI: -1.724, -0.612). No difference in the percentage of donor-reactive T-cells in the circulation at transplant biopsy was found between subjects experiencing a TCMR and the control group [p = 0.64 (CD4+), p = 0.52 (CD8+)]. Graft-infiltrating T-cells showed an up to six-fold increase of donor-reactive T-cell clonotypes compared to the blood at the same time (3.7 vs. 0.6% and 2.4 vs. 1.5%), but the infiltrating TCR repertoire was not reflected by the composition of the circulating TCR repertoire despite some overlap. Network analysis showed a distinct segregation of the donor-reactive repertoire with higher modularity than the overall TCR repertoire in the blood. These findings indicate an unchoreographed process of diverse T-cell clones directed against numerous non-self antigens found in the allograft. Conclusion: Donor-reactive T-cells are enriched in the kidney allograft during a TCMR episode, and dominant tissue clones are also found in the blood. Trial Registration: Clinicaltrials.gov: NCT: 03422224 (https://clinicaltrials.gov/ct2/show/NCT03422224).

Hypolipidemic effects of herbal extracts by reduction of adipocyte differentiation, intracellular neutral lipid content, lipolysis, fatty acid exchange and lipid droplet motility.

An increase in adipose tissue is caused by the increased size and number of adipocytes. Lipids accumulate in intracellular stores, known as lipid droplets (LDs). Recent studies suggest that parameters such as LD size, shape and dynamics are closely related to the development of obesity. Berberine (BBR), a natural plant alkaloid, has been demonstrated to possess anti-obesity effects. However, it remains unknown which cellular processes are affected by this compound or how effective herbal extracts containing BBR and other alkaloids actually are. For this study, we used extracts of Coptis chinensis, Mahonia aquifolium, Berberis vulgaris and Chelidonium majus containing BBR and other alkaloids and studied various processes related to adipocyte functionality. The presence of extracts resulted in reduced adipocyte differentiation, as well as neutral lipid content and rate of lipolysis. We observed that the intracellular fatty acid exchange was reduced in different LD size fractions upon treatment with BBR and Coptis chinensis. In addition, LD motility was decreased upon incubation with BBR, Coptis chinensis and Chelidonium majus extracts. Furthermore, Chelidonium majus was identified as a potent fatty acid uptake inhibitor. This is the first study that demonstrates the selected regulatory effects of herbal extracts on adipocyte function.

Data Based Prediction of Blood Glucose Concentrations Using Evolutionary Methods.

Predicting glucose values on the basis of insulin and food intakes is a difficult task that people with diabetes need to do daily. This is necessary as it is important to maintain glucose levels at appropriate values to avoid not only short-term, but also long-term complications of the illness. Artificial intelligence in general and machine learning techniques in particular have already lead to promising results in modeling and predicting glucose concentrations. In this work, several machine learning techniques are used for the modeling and prediction of glucose concentrations using as inputs the values measured by a continuous monitoring glucose system as well as also previous and estimated future carbohydrate intakes and insulin injections. In particular, we use the following four techniques: genetic programming, random forests, k-nearest neighbors, and grammatical evolution. We propose two new enhanced modeling algorithms for glucose prediction, namely (i) a variant of grammatical evolution which uses an optimized grammar, and (ii) a variant of tree-based genetic programming which uses a three-compartment model for carbohydrate and insulin dynamics. The predictors were trained and tested using data of ten patients from a public hospital in Spain. We analyze our experimental results using the Clarke error grid metric and see that 90% of the forecasts are correct (i.e., Clarke error categories A and B), but still even the best methods produce 5 to 10% of serious errors (category D) and approximately 0.5% of very serious errors (category E). We also propose an enhanced genetic programming algorithm that incorporates a three-compartment model into symbolic regression models to create smoothed time series of the original carbohydrate and insulin time series.

Single Molecule Fluorescence Microscopy and Machine Learning for Rhesus D Antigen Classification.

In transfusion medicine, the identification of the Rhesus D type is important to prevent anti-D immunisation in Rhesus D negative recipients. In particular, the detection of the very low expressed DEL phenotype is crucial and hence constitutes the bottleneck of standard immunohaematology. The current method of choice, adsorption-elution, does not provide unambiguous results. We have developed a complementary method of high sensitivity that allows reliable identification of D antigen expression. Here, we present a workflow composed of high-resolution fluorescence microscopy, image processing, and machine learning that - for the first time - enables the identification of even small amounts of D antigen on the cellular level. The high sensitivity of our technique captures the full range of D antigen expression (including D+, weak D, DEL, D-), allows automated population analyses, and results in classification test accuracies of up to 96%, even for very low expressed phenotypes.

Biomolecular Characterization of Putative Antidiabetic Herbal Extracts.

Induction of GLUT4 translocation in the absence of insulin is considered a key concept to decrease elevated blood glucose levels in diabetics. Due to the lack of pharmaceuticals that specifically increase the uptake of glucose from the blood circuit, application of natural compounds might be an alternative strategy. However, the effects and mechanisms of action remain unknown for many of those substances. For this study we investigated extracts prepared from seven different plants, which have been reported to exhibit anti-diabetic effects, for their GLUT4 translocation inducing properties. Quantitation of GLUT4 translocation was determined by total internal reflection fluorescence (TIRF) microscopy in insulin sensitive CHO-K1 cells and adipocytes. Two extracts prepared from purslane (Portulaca oleracea) and tindora (Coccinia grandis) were found to induce GLUT4 translocation, accompanied by an increase of intracellular glucose concentrations. Our results indicate that the PI3K pathway is mainly responsible for the respective translocation process. Atomic force microscopy was used to prove complete plasma membrane insertion. Furthermore, this approach suggested a compound mediated distribution of GLUT4 molecules in the plasma membrane similar to insulin stimulated conditions. Utilizing a fluorescent actin marker, TIRF measurements indicated an impact of purslane and tindora on actin remodeling as observed in insulin treated cells. Finally, in-ovo experiments suggested a significant reduction of blood glucose levels under tindora and purslane treated conditions in a living organism. In conclusion, this study confirms the anti-diabetic properties of tindora and purslane, which stimulate GLUT4 translocation in an insulin-like manner.

ImmunExplorer (IMEX): a software framework for diversity and clonality analyses of immunoglobulins and T cell receptors on the basis of IMGT/HighV-QUEST preprocessed NGS data.

BACKGROUND: Today's modern research of B and T cell antigen receptors (the immunoglobulins (IG) or antibodies and T cell receptors (TR)) forms the basis for detailed analyses of the human adaptive immune system. For instance, insights in the state of the adaptive immune system provide information that is essentially important in monitoring transplantation processes and the regulation of immune suppressiva. In this context, algorithms and tools are necessary for analyzing the IG and TR diversity on nucleotide as well as on amino acid sequence level, identifying highly proliferated clonotypes, determining the diversity of the cell repertoire found in a sample, comparing different states of the human immune system, and visualizing all relevant information. RESULTS: We here present IMEX, a software framework for the detailed characterization and visualization of the state of human IG and TR repertoires. IMEX offers a broad range of algorithms for statistical analysis of IG and TR data, CDR and V-(D)-J analysis, diversity analysis by calculating the distribution of IG and TR, calculating primer efficiency, and comparing multiple data sets. We use a mathematical model that is able to describe the number of unique clonotypes in a sample taking into account the true number of unique sequences and read errors; we heuristically optimize the parameters of this model. IMEX uses IMGT/HighV-QUEST analysis outputs and includes methods for splitting and merging to enable the submission to this portal and to combine the outputs results, respectively. All calculation results can be visualized and exported. CONCLUSION: IMEX is an user-friendly and flexible framework for performing clonality experiments based on CDR and V-(D)-J rearranged regions, diversity analysis, primer efficiency, and various different visualization experiments. Using IMEX, various immunological reactions and alterations can be investigated in detail. IMEX is freely available for Windows and Unix platforms at http://bioinformatics.fh-hagenberg.at/immunexplorer/.

Characterization of the Distance Relationship Between Localized Serotonin Receptors and Glia Cells on Fluorescence Microscopy Images of Brain Tissue.

We here present two new methods for the characterization of fluorescent localization microscopy images obtained from immunostained brain tissue sections. Direct stochastic optical reconstruction microscopy images of 5-HT1A serotonin receptors and glial fibrillary acidic proteins in healthy cryopreserved brain tissues are analyzed. In detail, we here present two image processing methods for characterizing differences in receptor distribution on glial cells and their distribution on neural cells: One variant relies on skeleton extraction and adaptive thresholding, the other on k-means based discrete layer segmentation. Experimental results show that both methods can be applied for distinguishing classes of images with respect to serotonin receptor distribution. Quantification of nanoscopic changes in relative protein expression on particular cell types can be used to analyze degeneration in tissues caused by diseases or medical treatment.

Identification of novel insulin mimetic drugs by quantitative total internal reflection fluorescence (TIRF) microscopy.

BACKGROUND AND PURPOSE: Insulin stimulates the transport of glucose in target tissues by triggering the translocation of glucose transporter 4 (GLUT4) to the plasma membrane. Resistance to insulin, the major abnormality in type 2 diabetes, results in a decreased GLUT4 translocation efficiency. Thus, special attention is being paid to search for compounds that are able to enhance this translocation process in the absence of insulin. EXPERIMENTAL APPROACH: Total internal reflection fluorescence (TIRF) microscopy was applied to quantify GLUT4 translocation in highly insulin-sensitive CHO-K1 cells expressing a GLUT4-myc-GFP fusion protein. KEY RESULTS: Using our approach, we demonstrated GLUT4 translocation modulatory properties of selected substances and identified novel potential insulin mimetics. An increase in the TIRF signal was found to correlate with an elevated glucose uptake. Variations in the expression level of the human insulin receptor (hInsR) showed that the insulin mimetics identified stimulate GLUT4 translocation by a mechanism that is independent of the presence of the hInsR. CONCLUSIONS AND IMPLICATIONS: Taken together, the results indicate that TIRF microscopy is an excellent tool for the quantification of GLUT4 translocation and for identifying insulin mimetic drugs.

Quantification and kinetic analysis of Grb2-EGFR interaction on micro-patterned surfaces for the characterization of EGFR-modulating substances.

The identification of the epidermal growth factor receptor (EGFR) as an oncogene has led to the development of several anticancer therapeutics directed against this receptor tyrosine kinase. However, drug resistance and low efficacy remain a severe challenge, and have led to a demand for novel systems for an efficient identification and characterization of new substances. Here we report on a technique which combines micro-patterned surfaces and total internal reflection fluorescence (TIRF) microscopy (µ-patterning assay) for the quantitative analysis of EGFR activity. It does not simply measure the phosphorylation of the receptor, but instead quantifies the interaction of the key signal transmitting protein Grb2 (growth factor receptor-bound protein 2) with the EGFR in a live cell context. It was possible to demonstrate an EGF dependent recruitment of Grb2 to the EGFR, which was significantly inhibited in the presence of clinically tested EGFR inhibitors, including small tyrosine kinase inhibitors and monoclonal antibodies targeting the EGF binding site. Importantly, in addition to its potential use as a screening tool, our experimental setup offers the possibility to provide insight into the molecular mechanisms of bait-prey interaction. Recruitment of the EGFR together with Grb2 to clathrin coated pits (CCPs) was found to be a key feature in our assay. Application of bleaching experiments enabled calculation of the Grb2 exchange rate, which significantly changed upon stimulation or the presence of EGFR activity inhibiting drugs.