A maximum-likelihood base caller for DNA sequencing

Classification of premalignant pancreatic cancer mass-spectrometry data using decision tree ensembles

Background

Pancreatic cancer is the fourth leading cause of cancer death in the United States. Consequently, identification of clinically relevant biomarkers for the early detection of this cancer type is urgently needed. In recent years, proteomics profiling techniques combined with various data analysis methods have been successfully used to gain critical insights into processes and mechanisms underlying pathologic conditions, particularly as they relate to cancer. However, the high dimensionality of proteomics data combined with their relatively small sample sizes poses a significant challenge to current data mining methodology where many of the standard methods cannot be applied directly. Here, we propose a novel methodological framework using machine learning method, in which decision tree based classifier ensembles coupled with feature selection methods, is applied to proteomics data generated from premalignant pancreatic cancer.

Results

This study explores the utility of three different feature selection schemas (Student t test, Wilcoxon rank sum test and genetic algorithm) to reduce the high dimensionality of a pancreatic cancer proteomic dataset. Using the top features selected from each method, we compared the prediction performances of a single decision tree algorithm C4.5 with six different decision-tree based classifier ensembles (Random forest, Stacked generalization, Bagging, Adaboost, Logitboost and Multiboost). We show that ensemble classifiers always outperform single decision tree classifier in having greater accuracies and smaller prediction errors when applied to a pancreatic cancer proteomics dataset.

Conclusion

In our cross validation framework, classifier ensembles generally have better classification accuracies compared to that of a single decision tree when applied to a pancreatic cancer proteomic dataset, thus suggesting its utility in future proteomics data analysis. Additionally, the use of feature selection method allows us to select biomarkers with potentially important roles in cancer development, therefore highlighting the validity of this method.

Improvement of base-calling in multilane automated DNA sequencing by use of electrophoretic calibration standards, data linearization, and trace alignment

We present a new method for the linearization and alignment of data traces generated by multilane automated DNA sequencing instruments. Application of this method to data generated with the Visible Genetics Open Gene DNA sequencing system (using MicroCel 700 gel cassettes, with a 25 cm separation distance) allows read lengths of > 1,000 nucleotides to be routinely obtained with high confidence and > 97% accuracy. This represents an increase of 10-15% in average read length, relative to data from this system that have not been processed in the fashion described herein. Most importantly, the linearization and alignment method allows usable sequence to be obtained from a fraction of 10-15% of data sets which, because of original trace misalignment problems, would otherwise have to be discarded. Our method involves adding electrophoretic calibration standards to the DNA sequencing fragments. The calibration standards are labeled with a dye that differs spectrally from the dye attached to the sequencing fragments. The calibration standards are identical in all the lanes. Analysis of the mobilities of the calibration standards allows correction for both systematic and random variation of electrophoretic properties between gel lanes. We have successfully used this method with two-dye and three-dye DNA sequencing instruments.