Archer® Analysis 6.0 highlights:
- HIPAA compliance
- Normal data sets for efficient noise characterization and confident mutation detection
- Error correction improvements for increased specificity
- UI improvements, including easy job unarchiving and selective file downloading
- More flexible than ever – preprocessing-only jobs, REST API + post script hooks for true customization
The Archer® Analysis bioinformatics platform has been updated to version 6.0 to provide more accurate variant and CNV calling, improved user experience and easy integration into existing workflows. While there are many small updates that improve performance and usability, here are a few key updates that you might be interested in.
Analysis Unlimited, the cloud-based configuration of Archer® Analysis, is third party verified HIPAA compliant, ensuring that all data remains protected and private at all times.
Addition of normal data sets for efficient noise characterization and confident mutation detection
While the concept of noise characterization and 95 MDAF (minimum detectable allele frequency by which a variant can be distinguished from the underlying noise at a probability of 0.95) was introduced in Analysis 5.0, this release adds some highly requested features to utilize the function efficiently and effectively. Archer® Analysis now supports the designation of historical normal tissue samples that can be used to calculate background noise from technical and biological variation without the need to reprocess the set in every run. These samples, termed normal data sets, strengthen the accuracy of the variant and CNV calls that Archer® Analysis makes.
Normal data sets are provided and designated by the user, who should ensure that specific samples do NOT contain the variations that the user wishes to measure in the test. If the specific variants are present, the ability to accurately estimate the background noise of those variations will be diminished, subsequently lowering the sensitivity of detecting those variations in tumor samples. Also, because it is not possible to acquire perfectly normal tissue, multiple samples from multiple sources should be designated to ensure that variants or SNPs that are specific to individual samples can be averaged out.
To designate a sample as a normal data set, first run the sample and select a target region. When the job is completed, go to Configuration–>Normal Data Sets and click the Create Normal Data Set button to designate the sample as a normal data set. Then, whenever you want to run a sample against the normal data set, just select the same target region as you are setting up the run and Archer® Analysis will give you the option to select the normal data set. The normal data set is not processed with the run but is rather used after analysis to compare to variant and CNV calls in the run.
Per-position p-values (defined as the probability that the variant is due to noise) are used in the variant grid. Within-run outlier detection can also be toggled to eliminate run-level noise and provide additional confidence.
With normal data sets, Archer® Analysis is bringing new confidence to mutation detection.
Under the hood improvements to error correction for increased specificity
We’ve learned quite a bit about using the molecular barcodes (MBCs) in Anchored Multiplex PCR (AMP™) chemistry for PCR and sequencing error correction. The Analysis 6.0 update features improvements to the algorithms that reduce false positives at very low allele fractions.
UI improvements, including easy job unarchiving and selective file downloading
Archiving old jobs is a great way to save space on your server or in the cloud. With Archer® Analysis, you can now unarchive old jobs in seconds right from the interface. Just find the archived job, click the dropdown arrow in the yellow Archived box, and click Unarchive. Your job is now ready for further review.
Need to download files from Archer® Analysis to plug into another program? Now you can download only the files that you want—no more sucking up internet bandwidth and hard drive space downloading all files associated with a specific job.
More flexible than ever: preprocessing-only jobs, plus REST API and job hooks for true customization
Version 6.0 brings more opportunities to integrate with and customize your Archer® Analysis pipeline. A robust API means that Archer® Analysis can be integrated into your existing lab management or variant interpretation platforms. Additionally, pre- and post-job hooks can automate processes before or after the job has completed. For example, create a pre-hook that converts lowercase letters in FASTQ files to uppercase before beginning a job. Or use a post hook to get a custom run QC report by email after a job is completed. All bits of data and file types can be used or repurposed with these flexible connectors. Analysis 6.0 also includes a preprocessing pipeline (MBC deduplication and error correction only), so that post-script hooks can tap into unaligned data for rapid non-human genome alignment for brand new AMP™ applications (e.g., gene editing, viral detection).