The development of many compounds must be discontinued because of severe side effects of some toxic metabolites, most often chemically reactive compounds. Of course, also potential drug candidates offer such a challenge to the metabolic system. In only a few generations, evolution did not have enough time to optimize the enzymes for this new challenge. The situation changed about two centuries ago, after the advent of synthetic organic compounds: many of them contain structural features that the metabolic system cannot handle in the same manner as natural products. Whereas many plant products are toxic, there are only rare examples that the metabolic system converts harmless natural substances into toxic entities. Modification, degradation, and/or conjugation, in many cases to polar products, enable a safe elimination from the organism. Over millions of years, a plethora of oxidizing, hydrolyzing, conjugating, and other enzymes were optimized by evolution. In addition to mediating cell metabolism, the metabolic system developed in animals and humans for the chemical conversion of xenobiotics. The software is available under GPL2+ from. Furthermore, the visualization capabilities enable exploratory analysis on the learnt dependencies and pave the way for a personalized prediction of phenotypes. We introduce a new method for robust prediction of phenotypes from molecular measurements such as DNA methylation or gene expression. This discovery supports the hypothesis that the ribosomes are a new frontier in genadaptivelearninge regulation. Applied to a leukemia data set, our method finds several ribosomal proteins associated with the risk group, which might be interesting targets for follow-up studies. Moreover, we show how to visualize the learned classifiers to display interesting associations with the target label. The new method achieves state-of-the-art performance on many different cancer data sets with measured DNA methylation or gene expression. In this way, the method provides a prediction that is adjusted for the potential biases on a per-patient basis, providing a personalized prediction for each test patient. We introduce a method that can estimate this bias on a per-feature level and incorporate calculated feature confidences into a weighted combination of classifiers with disjoint feature sets. This is especially true when these biases differ between the training and test set. If a model does not take potential biases in the data into account, this can lead to problems when trying to predict the stage of a certain cancer type. Furthermore, some cancer types are very heterogeneous, meaning that there might be different underlying causes for the same type of cancer among different individuals. This makes it harder to reproduce the exact values of measurements coming from different laboratories. Molecular measurements from cancer patients such as gene expression and DNA methylation can be influenced by several external factors.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |