Predictive Classification of IBS-subtype: Performance of a 250-gene RNA expression panel vs. Complete Blood Count (CBC) profiles under a Random Forest model.

In this experiment, an R-script was developed to select the best performing machine learning (ML) predictive classification algorithm for IBS subtype, and compare the performance of two datasets from the same clinical cohort: 1) The Complete Blood Count (CBC) results, and 2) A 250 gene Nanostring expression panel run on RNA from the Buffy Coat fraction. This publicly available data was compiled from open-source repositories and previously published supplementary data. Column labels were reformatted according to tidy-data standards. NA values in the data were imputed based on the mean value of the data column. Subject groups included Control (ie. healthy), IBS-D (diarrhea predominant), and IBS-C (constipation predominant) subtypes. These groups had unequal numbers in the original study, and so random re-sampling was used to make the group numbers equal for downstream linear regression-based analyses. The data was randomly split into training and validation subsets, and 5 classification algorithms were tested. Random Forest was clearly the best performing algorithm for both CBC and gene expression panel data, generally with >95% predictive accuracy, without additional tuning. The 250-gene RNA expression panel performed somewhat better than the CBC profile under a Random Forest model, however the CBC profiles had only 13 predictor variables vs. the 250 of the RNA expression panel. Some artifacts may result from the duplication of IBS-D and IBS-C rows from to the group-size balancing method, and so larger and more comprehensive datasets will be obtained for a follow-up analysis. The R-script and reformatted data are published as supplementary material here, and as a component of the AnalyzeBloodworkv1.2 GitHub repository.

A Reconciliation of Packed Column Permeability Data: Column Permeability as a Function of Particle Porosity

In his textbook teaching of packed bed permeability, Georges Guiochon uses mobile phase velocity as the fluid velocity term in his elaboration of the Darcy permeability equation. Although this velocity frame makes a lot of sense from a thermodynamic point of view, it is valid only with respect to permeability at a single theoretical boundary condition. In his more recent writings, however, Guiochon has departed from his long-standing mode of discussing permeability in terms of the Darcy equation and has embraced the well-known Kozeny-Blake equation. In this paper, his teaching pertaining to the constant in the Kozeny-Blake equation is examined and, as a result, a new correlation coefficient is identified and defined herein based on the velocity frame used in his teaching. This coefficient correlates pressure drop and fluid velocity as a function of particle porosity. We show that in their experimental protocols, Guiochon et al. have not adhered to a strict material balance of permeability which creates a mismatch of particle porosity and leads to erroneous conclusions regarding the value of the permeability coefficient in the Kozeny-Blake equation. By correcting the experimental data to properly reflect particle porosity we reconcile the experimental results of Guiochon and Giddings, resulting in a permeability reference chart which is presented here for the first time. This reference chart demonstrates that Guiochon’s experimental data, when properly normalized for particle porosity and other related discrepancies, corroborates the value of 267 for the constant in the Kozeny-Blake equation which was derived by Giddings in 1965.