Targeted sequencing of PCR amplicons generated from bisulfite deaminated DNA is a flexible, cost-effective way to study methylation of a sample at single CpG resolution and perform subsequent multi-target, multi-sample comparisons. Currently, no platform specific protocol, support, or analysis solution is provided to perform targeted bisulfite sequencing on a Personal Genome Machine (PGM). Here, we present a novel tool, called TABSAT, for analyzing targeted bisulfite sequencing data generated on Ion Torrent sequencers. The workflow starts with raw sequencing data, performs quality assessment, and uses a tailored version of Bismark to map the reads to a reference genome. The pipeline visualizes results as lollipop plots and is able to deduce specific methylation-patterns present in a sample. The obtained profiles are then summarized and compared between samples. In order to assess the performance of the targeted bisulfite sequencing workflow, 48 samples were used to generate 53 different Bisulfite-Sequencing PCR amplicons from each sample, resulting in 2,544 amplicon targets. We obtained a mean coverage of 282X using 1,196,822 aligned reads. Next, we compared the sequencing results of these targets to the methylation level of the corresponding sites on an Illumina 450k methylation chip. The calculated average Pearson correlation coefficient of 0.91 confirms the sequencing results with one of the industry-leading CpG methylation platforms and shows that targeted amplicon bisulfite sequencing provides an accurate and cost-efficient method for DNA methylation studies, e.g., to provide platform-independent confirmation of Illumina Infinium 450k methylation data. TABSAT offers a novel way to analyze data generated by Ion Torrent instruments and can also be used with data from the Illumina MiSeq platform. It can be easily accessed via the Platomics platform, which offers a web-based graphical user interface along with sample and parameter storage. TABSAT is freely available under a GNU General Public License version 3.0 (GPLv3) at https://github.com/tadkeys/tabsat/ and http://demo.platomics.com/.

Anomaly detection in large industrial cooling systems is very challenging due to the high data dimensionality, inconsistent sensor recordings, and lack of labels. The state of the art for automated anomaly detection in these systems typically relies on expert knowledge and thresholds. However, data is viewed isolated and complex, multivariate relationships are neglected. In this work, we present an autoencoder based end-to-end workflow for anomaly detection suitable for multivariate time series data in large industrial cooling systems, including explained fault localization and root cause analysis based on expert knowledge. We identify system failures using a threshold on the total reconstruction error (autoencoder reconstruction error including all sensor signals). For fault localization, we compute the individual reconstruction error (autoencoder reconstruction error for each sensor signal) allowing us to identify the signals that contribute most to the total reconstruction error. Expert knowledge is provided via look-up table enabling root-cause analysis and assignment to the affected subsystem. We demonstrated our findings in a cooling system unit including 34 sensors over a 8-months’ time period using 4-fold cross validation approaches and automatically created labels based on thresholds provided by domain experts. Using 4-fold cross validation, we reached a F1-score of 0.56, whereas the autoencoder results showed a higher consistency score (CS of 0.92) compared to the automatically created labels (CS of 0.62) – indicating that theanomaly is recognized in a very stable manner. The automatically created labels, however, detected anomaly earlier. The main anomaly was found by the autoencoder and automatically created labels, and was also recorded in the log files. Further, the explained fault localization highlighted the most affected component for the main anomaly in a very consistent manner.

No abstract

Anomaly detection in large industrial cooling systems is very challenging due to the high data dimensionality, inconsistent sensor recordings, and lack of labels. The state of the art for automated anomaly detection in these systems typically relies on expert knowledge and thresholds. However, data is viewed isolated and complex, multivariate relationships are neglected. In this work, we present an autoencoder based end-to-end workflow for anomaly detection suitable for multivariate time series data in large industrial cooling systems, including explained fault localization and root cause analysis based on expert knowledge. We identify system failures using a threshold on the total reconstruction error (autoencoder reconstruction error including all sensor signals). For fault localization, we compute the individual reconstruction error (autoencoder reconstruction error for each sensor signal) allowing us to identify the signals that contribute most to the total reconstruction error. Expert knowledge is provided via look-up table enabling root-cause analysis and assignment to the affected subsystem. We demonstrated our findings in a cooling system unit including 34 sensors over a 8-months' time period using 4-fold cross validation approaches and automatically created labels based on thresholds provided by domain experts. Using 4-fold cross validation, we reached a F1-score of 0.56, whereas the autoencoder results showed a higher consistency score (CS of 0.92) compared to the automatically created labels (CS of 0.62) -indicating that the Stephanie Holly et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.anomaly is recognized in a very stable manner. The automatically created labels, however, detected anomaly earlier. The main anomaly was found by the autoencoder and automatically created labels, and was also recorded in the log files. Further, the explained fault localization highlighted the most affected component for the main anomaly in a very consistent manner.