Jupyter Notebook V-PFCRC - Formula Generation and Validation

This Jupyter Notebook contains comprehensive Python scripts designed to automate the generating and validating V-PFCRC (Volume- and Creatinine-corrected) formulas. These scripts are intended to analyze data from the accompanying Excel file, "Arsenic Raw Data Set 1.xlsx." However, they can easily be adapted to any Excel file that includes creatinine (CRN) values and uncorrected spot urinary analyte data. Modifications to file names, paths, and analyte names within the scripts will be necessary for use with a different dataset.The notebook automates the following key tasks for V-PFCRC formula generation:<b>Generation of V-PFCRC Coefficients</b>: It calculates analyte-specific, sex-differentiated V-PFCRC coefficients based on input datasets.<b>CRN Cut-off Optimization</b>: The notebook identifies the optimal creatinine range for V-PFCRC by evaluating regression curve fits across a broad spectrum of creatinine cut-offs.<b>Comparison of Mean Calculation Methods</b>: It compares different methods for mean calculation (geometric mean, arithmetic mean, and median) and evaluates their effectiveness in improving regression model fits.<b>Validation of the V-PFCRC Formula</b>: The performance of the V-PFCRC formula is validated by comparing residual R² values across datasets that include uncorrected data, conventionally creatinine-corrected data, and V-PFCRC-adjusted data.Additionally, the accompanying Excel file, "Jupyter-Python V-PFCRC Formula Generation Validation Results.xlsx," provides critical regression data used for optimizing and validating the V-PFCRC formulas. Beyond arsenic, four smaller datasets containing data for metals such as Cesium, Molybdenum, Strontium, and Zinc were also analyzed, along with a large dataset for Iodine. The results from these analyses are provided in the file.<br>