A data-driven acute inflammation therapy

Acute inflammation is a severe medical condition defined as an inflammatory response of the body to an infection. Its rapid progression requires quick and accurate decisions from clinicians. Inadequate and delayed decisions makes acute inflammation the 10th leading cause of death overall in United States with the estimated cost of treatment about $17 billion annually. However, despite the need, there are limited number of methods that could assist clinicians to determine optimal therapies for acute inflammation. We developed a data-driven method for suggesting optimal therapy by using machine learning model that is learned on historical patients' behaviors. To reduce both the risk of failure and the expense for clinical trials, our method is evaluated on a virtual patients generated by a mathematical model that emulates inflammatory response. In conducted experiments, acute inflammation was handled with two complimentary pro- and anti-inflammatory medications which adequate timing and doses are crucial for the successful outcome. Our experiments show that the dosage regimen assigned with our data-driven method significantly improves the percentage of healthy patients when compared to results by other methods used in clinical practice and found in literature. Our method saved 88% of patients that would otherwise die within a week, while the best method found in literature saved only 73% of patients. At the same time, our method used lower doses of medications than alternatives. In addition, our method achieved better results than alternatives when only incomplete or noisy measurements were available over time as well as it was less affected by therapy delay. The presented results provide strong evidence that models from the artificial intelligence community have a potential for development of personalized treatment strategies for acute inflammation.     

Molecular analysis of Leptospira spp. isolated from humans by restriction fragment length polymorphism, real-time PCR and pulsed-field gel electrophoresis

A total of 17 Leptospira clinical strains isolated from humans in Croatia were serologically and genetically analysed. For serovar identification, the microscopic agglutination test (MAT) and pulsed-field gel electrophoresis (PFGE) were used. To identify isolates on genomic species level, PCR-based restriction fragment length polymorphism (RFLP) and real-time PCR were performed. MAT revealed the following serogroup affinities: Grippotyphosa (seven isolates), Icterohaemorrhagiae (eight isolates) and Javanica (two isolates). RFLP of PCR products from a 331-bp-long fragment of rrs (16S rRNA gene) digested with endonucleases MnlI and DdeI and real-time PCR revealed three Leptospira genomic species. Grippotyphosa isolates belonged to Leptospira kirschneri , Icterohaemorrhagiae isolates to Leptospira interrogans and Javanica isolates to Leptospira borgpetersenii . Genomic DNA from 17 leptospiral isolates was digested with NotI and SgrAI restriction enzymes and analysed by PFGE. Results showed that seven isolates have the same binding pattern to serovar Grippotyphosa, eight isolates to serovar Icterohaemorrhagiae and two isolates to serovar Poi. Results demonstrate the diversity of leptospires circulating in Croatia. We point out the usefulness of a combination of PFGE, RFLP and real-time PCR as appropriate molecular methods in molecular analysis of leptospires.     

Characterization of maximal enzyme catalytic rates in central metabolism of Arabidopsis thaliana

Availability of plant‐specific enzyme kinetic data is scarce, limiting the predictive power of metabolic models and precluding identification of genetic factors of enzyme properties. Enzyme kinetic data are measured in vitro, often under non‐physiological conditions, and conclusions elicited from modeling warrant caution. Here we estimate maximal in vivo catalytic rates for 168 plant enzymes, including photosystems I and II, cytochrome‐b6f complex, ATP‐citrate synthase, sucrose‐phosphate synthase as well as enzymes from amino acid synthesis with previously undocumented enzyme kinetic data in BRENDA. The estimations are obtained by integrating condition‐specific quantitative proteomics data, maximal rates of selected enzymes, growth measurements from Arabidopsis thaliana rosette with and fluxes through canonical pathways in a constraint‐based model of leaf metabolism. In comparison to findings in Escherichia coli, we demonstrate weaker concordance between the plant‐specific in vitro and in vivo enzyme catalytic rates due to a low degree of enzyme saturation. This is supported by the finding that concentrations of nicotinamide adenine dinucleotide (phosphate), adenosine triphosphate and uridine triphosphate, calculated based on our maximal in vivo catalytic rates, and available quantitative metabolomics data are below reported values and, therefore, indicate undersaturation of respective enzymes. Our findings show that genome‐wide profiling of enzyme kinetic properties is feasible in plants, paving the way for understanding resource allocation.     

Total phenolic content, flavonoid concentration, antioxidant and antimicrobial activity of methanol extracts from three Seseli L. taxa

The present study describes the total phenolic content, concentrations of flavonoids and in vitro antioxidant and antimicrobial activity of methanol extracts from Seseli pallasii Besser, S. libanotis (L.) Koch ssp. libanotis and S. libanotis (L.) Koch ssp. intermedium (Rupr.) P. W. Ball, growing wild in Serbia. The total phenolic content in the extracts was determined using Folin-Ciocalteu reagent and their amounts ranged between 84.04 to 87.52 mg GA (gallic acid)/g. The concentrations of flavonoids in the extracts varied from 4.75 to 19.37 mg Qu (quercetin)/g. Antioxidant activity was analyzed using DPPH reagent. Antioxidant activity ranged from 0.46 to 4.63 IC₅₀ (mg/ml) and from 1.98 to 2.19 mg VitC (vitamin C)/g when tested with the DPPH and ABTS reagents, respectively, using BHA and VitC as controls. The antimicrobial activity of the extracts was investigated using a micro-well dilution assay for the most common human gastrointestinal pathogenic bacterial strains: Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 9027, Salmonella enteritidis ATCC 13076, Bacillus cereus ATCC 10876, Listeria monocytogenes ATCC15313, Staphylococcus aureus ATCC 25923 and Candida albicans ATCC 10231. This finding suggests that Seseli species may be considered as a natural source of antioxidants and antimicrobial agents.     

Early classification of multivariate temporal observations by extraction of interpretable shapelets

ABSTRACT: BACKGROUND: Early classification of time series is beneficial for biomedical informatics problems suchincluding, but not limited to, disease change detection. Early classification can be oftremendous help by identifying the onset of a disease before it has time to fully take hold. Inaddition, extracting patterns from the original time series helps domain experts to gaininsights into the classification results. This problem has been studied recently using timeseries segments called shapelets. In this paper, we present a method, which we callMultivariate Shapelets Detection (MSD), that allows for early and patient-specificclassification of multivariate time series. The method extracts time series patterns, calledmultivariate shapelets, from all dimensions of the time series that distinctly manifest thetarget class locally. The time series were classified by searching for the earliest closestpatterns. RESULTS: The proposed early classification method for multivariate time series has been evaluated oneight gene expression datasets from viral infection and drug response studies in humans. Inour experiments, the MSD method outperformed the baseline methods, achieving highlyaccurate classification by using as little as 40%-64% of the time series. The obtained resultsprovide evidence that using conventional classification methods on short time series is notas accurate as using the proposed methods specialized for early classification. CONCLUSION: For the early classification task, we proposed a method called Multivariate ShapeletsDetection (MSD), which extracts patterns from all dimensions of the time series. Weshowed that the MSD method can classify the time series early by using as little as40%-64% of the time series' length.     

Influence of Circadian Light-Dark Alternations on Macrophages and Lymphocytes of CBA Mouse

The influence of circadian 12 h light-12 h dark alternations on CBA mouse macrophages and lymphocytes was determined using tests for macrophage spreading and ingestion ability or flow cytometry immunophenotyping of blood, lymph node, and spleen lymphocytes. The animals were tested every 4 h around the clock. Collected macrophages were incubated in vitro for 3 or 18 h. Monoclonal antibodies permitted detection of T-lymphocytes, suppressor-cytotoxic T-lymphocytes, helper-inducer T-lymphocytes, or B-lymphoeytes. Two types of analyses were performed: First, the difference between the same intervals of the 12 h light or dark period was determined. The macrophage ingestion was significantly lower at the beginning and higher at the end of the dark period. We have also found a significant increase in blood T-lymphocytes of helper-inducer T-lymphocyte percentages and of the T helper-inducenT suppressor-cytotoxic ratio during the dark period. Second, the ultradian variation during the 12 h light or dark period was determined. The variability was significant both for macrophage spreading and ingestion. Multiple significant variations of lymph node, spleen, or blood lymphocyte percentages were also observed. All of these data indicate that daily alteration of the lighting regimen significantly influences mouse peritoneal macrophage functions and various lymphocyte subsets.