Prediction of human pharmacokinetics based on preclinical in vitro and in vivo data.
Explore the source record for details and available documents.
Drug Design↗
Biomedical subjects
J Zuegge
Publications and source records attributed to J Zuegge.
Deciphering apicoplast targeting signals--feature extraction from nuclear-encoded precursors of Plasmodium falciparum apicoplast proteins.
The malaria causing protozoan Plasmodium falciparum contains a vestigal, non-photosynthetic plastid, the apicoplast. Numerous proteins encoded by nuclear genes are targeted to the apicoplast courtesy of N-terminal extensions. With the impending sequence completion of an entire genome of the malaria parasite, it is important to have software tools in place for prediction of subcellular locations for all proteins. Apicoplast targeting signals are bipartite; containing a signal peptide and a transit peptide. Nuclear-encoded apicoplast protein precursors were analyzed for characteristic features by statistical methods, principal component analysis, self-organizing maps, and supervised neural networks. The transit peptide contains a net positive charge and is rich in asparagine, lysine, and isoleucine residues. A novel prediction system (PATS, predict apicoplast-targeted sequences) was developed based on various sequence features, yielding a Matthews correlation coefficient of 0.91 (97% correct predictions) in a 40-fold cross-validation study. This system predicted 22% apicoplast proteins of the 205 potential proteins on P. falciparum chromosome 2, and 21% of 243 chromosome 3 proteins. A combination of the PATS results with a signal peptide prediction yields 15% potentially nuclear-encoded apicoplast proteins on chromosomes 2 and 3. The prediction tool will advance P. falciparum genome analysis, and it might help to identify apicoplast proteins as drug targets for the development of novel anti-malaria agents.
Algorithms↗
H-BloX: visualizing alignment block entropies.
H-BloX is a web-based JavaScript application that allows the calculation and visualization of Shannon information content or relative entropy (Kullback-Leibler 'distance') within sequence alignment blocks. The application was designed for use in both teaching and research. Amino acid, nucleic acid sequences, or any other type of aligned chemical structures may serve as the input. Various interpretations of the meaning of 'entropy' or 'information content' are possible, including treatment as a chemical diversity measure or the degree of feature conservation. For analysis of numerical data by H-BloX, values must be converted to a user-defined character alphabet before computation of entropy or information content. H-BloX was successfully applied to feature identification in Escherichia coli signal peptides and their cleavage sites. Characteristics known features became visible, e.g., the hydrophobic core region and the well-known '-3,-1' cleavage site pattern. Based on the H-BloX analysis, the hydrophobic core is centered at amino acid residue position 13, counting from the N-terminal end of the protein precursor sequence. This result was obtained by using a built-in feature of H-BloX that enables conversion of amino acid sequences to a different alphabet that is based on hydrophobicity assignments. H-BloX can be accessed online or downloaded as HTML/JavaScript at http://bopwww.biologie.uni-freiburg.de/~bioinfo/HBloX/html/index.html.
Bacterial Proteins↗
Prediction of hepatic metabolic clearance: comparison and assessment of prediction models.
OBJECTIVE: To perform a comparative quantitative evaluation of the prediction accuracy for human hepatic metabolic clearance of 5 different mathematical models: allometric scaling (multiple species and rat only), physiologically based direct scaling, empirical in vitro-in vivo correlation, and supervised artificial neural networks. METHODS: The mathematical prediction models were implemented with a publicly available dataset of 22 extensively metabolised compounds and compared for their prediction accuracy using 3 quality indicators: prediction error sum of squares (PRESS), r2 and the fold-error. RESULTS: Approaches such as physiologically based direct scaling, empirical in vitro-in vivo correlation and artificial neural networks, which are based on in vitro data only, yielded an average fold-error ranging from 1.64 to 2.03 and r2 values greater than 0.77, as opposed to r2 values smaller than 0.44 when using allometric scaling combining in vivo and in vitro preclinical data. The percentage of successful predictions (less than 2-fold error) ranged from 55% (rat allometric scaling) to between 64 and 68% with the other approaches. CONCLUSIONS: On the basis of a diverse set of 22 metabolised drug molecules, these studies showed that the most cost-effective and accurate approaches, such as physiologically based direct scaling and empirical in vitro-in vivo correlation, are based on in vitro data alone. Inclusion of in vivo preclinical data did not significantly improve prediction accuracy; the prediction accuracy of the allometric approaches was at the lower end of all methods compared.
Animals↗