br Acknowledgements Lavender Yuen Nam Fan is

Acknowledgements Lavender Yuen-Nam Fan is a PhD student and research assistant supported by Imperial College IC Trust. Paula Saavedra-García is a post-doctoral research associate supported by the Medical Research Council (MRC) of UK (MR/N012097/1). Eric W.-F. Lam׳s work is supported by MRC (MR/N012097/1), CRUK (A12011) and Breast Cancer Now (2012MayPR070; 2012NovPhD016).
Data Here we provide the curated human dataset for hyperbilirubinemia that was used for generating a classification model for hyperbilirubinemia [1]. Hyperbilirubinemia is the pathological condition of accumulation of conjugated or unconjugated bilirubin in sinusoidal blood. It has been associated with underlying liver disease, since it neuraminidase inhibitor is often accompanying severe liver conditions like hepatocellular drug induced liver injury [2–5] and cholestasis [6,7]. There is evidence in literature that there is causality between OATP1B1 and 1B3 inhibition and hyperbilirubinemia [8]. Apart from the hyperbilirubinemia data SMILES, we are also providing the calculated predictions of OATP1B1 and 1B3 inhibition obtained from models generated in a previous study [9]. The transporters predictions had been used as descriptors for the generation of the hyperbilirubinemia classification model. Furthermore, we provide the matrix of calculated MOE 2D descriptors.
Experimental design, materials and methods
Descriptors calculation All 2D MOE descriptors were calculated with the software MOE 2014.09 and out of 192 descriptors, 92 comprehensible descriptors were preselected without significant loss in the resulted hyperbilirubinemia model׳s performance (Table S1). Due to the evidence in literature for a link of OATP1B1 and 1B3 inhibition with hyperbilirubinemia, the predictions of OATP1B1 and 1B3 inhibition were also used as descriptors. The predictions were obtained using the OATP1B1 and 1B3 models generated in a previous study [9]. These models showed excellent accuracy and prospective predictivity, thus correctly identifying 9/10 compounds for OATP1B1 and 8/10 compounds for OATP1B3 in a virtual screening of DrugBank followed by experimental verification of the top ranked hits. For the human hyperbilirubinemia data 829/835 compounds (i.e 99.3% of the dataset) are within the applicability domain of both OATP1B1 and 1B3 models. The applicability domain was checked on KNIME with the Enalos nodes [15,16] that compute the applicability domain on the basis of the Euclidean distances [17].
Acknowledgments
We are thankful to ChemAxon (https://www.chemaxon.com/) for providing us with an Academic License of Marvin Suite. Marvin was used for drawing, displaying and characterizing chemical structures, substructures and reactions, Marvin 6.1.3., 2013, ChemAxon (http://www.chemaxon.com).
Data Table 1 presents the morphoagronomic characteristics of the cowpea seeds. Table 2 illustrates the structures of the 30 compounds identified in cowpea seeds with the corresponding 1H and 13C NMR chemical shifts, multiplicity, and constant coupling [4–8,10]. PC1 vs. PC3 scores and loadings coordinate system for different cultivars of cowpea evaluating only the aromatic region are presented in Fig. 1. Figs. 2 and 3 show the comparison of the 13C CP-MAS and the 13C SP-MAS spectra of the cowpea seeds [3].
Experimental design, materials and methods Fig. 4 presents nine cowpea seeds from the germplasm bank of the Center of Agricultural Science at Federal University of Ceará (CCA/UFC), Brazil, with the accession numbers and the vintage years.
Data Here, we provided BET isotherms of EDA, TETA and PEHA polymers in Fig. 1. TG, TGA graphics of EDA and TETA polymer are provided inFig. 2. The FTIR of the fresh, Au(III) adsorbed and stripped EDA, TETA and PEHA particles at wave numbers from 600 to 4000cm−1 are shown in Fig. 3.
BET isotherms See Fig. 1.
Thermal degradation results