46 compounds and their Observed logPapp values have been given in paper. "L-alanine" was retrieved as "Alanine", "AZT" as "Zidovudine", "L-Dopa" as "Levodopa" and "Methyl scopolamine" as "N-Methyl scopolamine" from ChemIDplus.

"The data set consists of 85 compounds, for which Wessel and co-workers gathered %HIA values from the literature. Sixteen of these 85 structures have Caco-2 cell permeability data, originating from Artursson and Karlsson that were used in the QSAR models proposed by Van de Waterbeemd et. al. and by Norinder et. al. In addition, Yazdanian et. al. reported Caco-2 cell permeability data for 29 of these 85 structures. Thus, three experimental measures were used to derive our QSAR model".

This URL links to the dataset on QSAR world Compounds and their property with respect to crossing of the "Blood-Brain Barrier" (BBB) have been given in the paper. In all 80 compounds with their BBB permabilty have been given in the files provided here. 45 "Cross" the BBB while 35 "Do Not Cross" the BBB

131 compounds and their 'Fraction Absorbed' values have been obtained from the table given in the paper. Some compounds had multiple values reported (as obtained from various references). These have been averaged and provided by us in the SDF and AMP files. The "Sanghvi_values.txt" file contains all data (as reported in Table 1) and "Sanghvi_avg_values.txt" contains averaged data for compounds with multiple entries (as reported in Table 2). "Benserazide, HCl" was retrieved as "Benserazide", "Ceftriaxone, Na" as "Ceftriaxone", "Foscarnet, Na" as "Foscarnet", "L-Dopa" as "Levodopa", "L-Leucine" as "Leucine" and "Triacrilast" as "Tiacrilast". The last compound was found to be mispelt as checked from the original reference.

A total of 86 drugs and their %HIA (Human Intestinal Absorption) values taken from the table given in the paper. Compound "trovoflaxicin" was retrieved as "trovafloxacin", "acetylsalicylic acid" as "aspirin", and "phenoxymethylpenicillinic acid" as "penicillinv" from ChemIDplus. Fragment removed from "timolol maleate" during "wash". AMP file has a "Label" column indicating 9 compounds used as a "Cross-Validation Set" and 10 compounds used as an "External Prediction Set" by the authors. Rest of the compounds are labeled as "Training Set".

51 compounds and their % Absorbed values have been provided in the paper. For some compounds the values have either not been given or indicated as irregular/poor/erratic/rapidly metabolized. THe datasets are available from QSAR World: http://www.qsarworld.com/qsar-datasets-yazdanian.php