Selective Partitioned Regression for Accurate Kidney Health Monitoring.

The number of people diagnosed with advanced stages of kidney disease have been rising every year. Early detection and constant monitoring are the only minimally invasive means to prevent severe kidney damage or kidney failure. We propose a cost-effective machine learning-based testing system that can facilitate inexpensive yet accurate kidney health checks. Our proposed framework, which was developed into an iPhone application, uses a camera-based bio-sensor and state-of-the-art classical machine learning and deep learning techniques for predicting the concentration of creatinine in the sample, based on colorimetric change in the test strip. The predicted creatinine concentration is then used to classify the severity of the kidney disease as healthy, intermediate, or critical. In this article, we focus on the effectiveness of machine learning models to translate the colorimetric reaction to kidney health prediction. In this setting, we thoroughly evaluated the effectiveness of our novel proposed models against state-of-the-art classical machine learning and deep learning approaches. Additionally, we executed a number of ablation studies to measure the performance of our model when trained using different meta-parameter choices. Our evaluation results indicate that our selective partitioned regression (SPR) model, using histogram of colors-based features and a histogram gradient boosted trees underlying estimator, exhibits much better overall prediction performance compared to state-of-the-art methods. Our initial study indicates that SPR can be an effective tool for detecting the severity of kidney disease using inexpensive lateral flow assay test strips and a smart phone-based application. Additional work is needed to verify the performance of the model in various settings.

Protein kinase C delta is a substrate of tissue transglutaminase and a novel autoantigen in coeliac disease.

Post-translational modification of proteins by deamidation or transamidation by tissue transglutaminase (tTG) has been suggested as a possible mechanism for the development of autoimmunity. Sequence analysis of protein kinase C delta (PKCδ) identified an amino acid motif that suggested the possibility that PKCδ was a glutamine substrate of tTG and MALDI-TOF analysis of synthesised peptides from PKCδ proved that this was the case. Polymerisation experiments using recombinant tTG and biotinylated hexapeptide substrate incorporation assays demonstrated that PKCδ is a substrate for tTG-mediated transamidation. Elevated levels of anti-PKCδ antibodies were detected in sera from patients with coeliac disease (p<0.0001) but not from patients with other autoimmune disorders. These data suggest that a subset of patients with coeliac disease produce autoantibodies against PKCδ and that this response may stem from a tTG-PKCδ substrate interaction.

Mapping of immune responses following wild-type and mutant ABeta42 plasmid or peptide vaccination in different mouse haplotypes and HLA Class II transgenic mice.

Although the recent clinical trial of the ABeta42 peptide vaccine against Alzheimer's Disease (AD) has been halted due to adverse events, the apparent clinical utility of this approach underscores the need to further improve the safety of the vaccine, as well as to understand the potential immunological basis for complications. In this study, we examine both humoral and cellular immune responses elicited by immunization with peptide or DNA encoding wild-type and the Flemish and Dutch mutations of ABeta42 (i.e. the beta amyloid peptide spanning amino acids 1-42) in mice of different immune haplotypes as well as HLA Class II transgenic mice. The Flemish and Dutch mutations have been associated with cerebrovascular hemorrhages in affected individuals. These data allow determination of potential immunological responses that could mediate pathology observed with mutant forms of amyloid beta, as well as lead to the generation of safer vaccine preparations. Following peptide or plasmid immunization, antibody responses were measured against the different ABeta42 peptides in an ELISA assay, while T cell epitopes were analyzed through interferon gamma ELISPOT and lymphocyte proliferation assays. B cell mapping studies indicated that sera from all of the haplotype mice vaccinated with any of the ABeta42 peptides reacted specifically to the first 10 amino acids of ABeta42 with the ABeta42 mutants eliciting higher immune responses. ELISPOT analysis, which accessed cellular immune responses indicated that mice expressed differences in Class I epitopes dependent on the different immune haplotypes. These results may have implications for the design of future ABeta42 based vaccines against Alzheimer's Disease.

Human duodenal epithelial cells constitutively express molecular components of antigen presentation but not costimulatory molecules.

Constitutive expression of major histocompatibility complex (MHC) class II molecules by duodenal epithelial cells (EC) suggests that they can present antigen to CD4(+) T cells. However, other molecular components including invariant chain (Ii), HLA-DM, and costimulatory molecules CD80, CD86 and CD40, are required for efficient T-cell activation. We have investigated whether normal human duodenal EC possess these molecules and whether they can mediate MHC class II antigen presentation. EC were isolated from duodenal biopsies from patients in whom pathology was excluded. Freshly-isolated duodenal EC did not stimulate autologous T-cell proliferation against purified protein derivative of tuberculin. Flow cytometry and immunoblot analysis revealed that duodenal EC constitutively express HLA-DR, Ii, and HLA-DM. Surface MHC class II associated invariant chain peptide (CLIP) was not detectable, suggesting that HLA-DM functions normally in CLIP removal. Duodenal EC expressed SDS-stable HLA-DR alphabeta heterodimers, indicating that peptide binding had occurred. Surface expression of CD80, CD86 or CD40 was not detected although mRNA for these costimulatory molecules was present in all samples. These results suggest that nondiseased human duodenal EC can process and present antigen by the MHC class II pathway, but that they may induce anergy, rather than activation, of local T cells.