ABSTRACT
Chemical reaction networks (CRNs) have been proposed as an abstraction for molecular computing. DNA strand displacement (DSD) reactions are good candidates to realize this endeavor, since DNA strands can be wired to implement the desired dynamic behavior in a test tube. Specialists use simulators to help them design such chemical systems before experimental implementation. In this sense, we present the DNAr package, an alternative open-source tool, developed in R language, for users from multidisciplinary areas. The current version of our tool offers functions to simulate CRNs, convert a formal CRN into a DSD network, interpret results, export to Visual DSD, and create libraries. Here, we use the consensus CRN to show DNAr features and a neural network model to demonstrate scalability, simulating more than 600 chemical reactions in a few minutes.
Subject(s)
DNA/metabolism , Software , Algorithms , Computer SimulationABSTRACT
A multivariate calibration model (PLS) was developed for the simultaneous spectrophotometric determination of Al(III) and Fe(III) in post-hemodialysis fluids with pyrocathecol violet (PCV) as chromogenic reagent. The analytes build stable complexes with PCV in presence of hexamine buffered medium at pH 6.1. The complexes show overlapped absorption bands in the spectral range of 220-800nm so that absorptions of 580 wavelengths were necessary for the calibrations. Determinations of Al(III) and Fe(III) were done without masking agents. The best calibration model was obtained by using PLS-1 regression with three components after data mean centering. The spectrophotometric method applied to assay the analytes in real post-hemodialysis samples containing no desferrioxamine B presented good agreement with voltammetric measurements used as reference. Concentrations ranging from 0.20 to 0.60mgL(-1) for Al(III) and for Fe(III) were determined in real samples. The multivariate detection limits for Al(III) and Fe(III) were 0.044 and 0.052mgL(-1), respectively, and the calculated values of sensitivity were 6.33 for Al(III) and 3.44 for Fe(III). The proposed method showed to be straightforward and useful to follow the hemodialysis progress for patients under treatment. Interferents were also investigated.
ABSTRACT
Second-order calibration and multivariate spectroscopic-kinetic measurements in the visible region are proposed to improve the Jaffé method for creatinine assay. Analyses performed on synthetic mixtures containing bilirubin, glucose, and albumin confirm that second-order calibration is useful for creatinine determination in human serum. Quantitative determinations of creatinine with the parallel factor analysis (PARAFAC) and direct trilinear decomposition (TLD) methods were compared. It is shown that both methods can be used for creatinine determination in human serum, with an SEP (standard error of prediction) of 2.22 and coefficient of variability of 6.14% for PARAFAC, and an SEP of 2.38 and coefficient of variability of 6.57% for TLD [corrected].
