A Molecular Barcode is a unique 8-bp sequence that is part of the MBC adapter and is used to identify unique fragments and “de-duplicate” the sequencing reads from a sample. This, along with the random start sites, helps identify and remove PCR duplicates.
Molecular Barcodes are an important element of Archer® Anchored Multiplex PCR (AMP™) chemistry. Watch the video below to learn more about how molecular barcodes work.
Please contact technical support at firstname.lastname@example.org to request a link to download the Archer Analysis Virtual Machine Installation Guide and/or User Guide. Additionally, informational videos and links to other supporting materials can be found at Videos Section.
If you already have the virtual machine or a login to our demo site, you can find the documentation under the Documents page.
Users must have a virtualization application, such as Virtual Box for single sample processing, or VMware for multiple concurrent sample processing, installed on their system first. The system must have at least 12GB of free RAM (beyond whatever else the host itself is using) and enough disk space to install the virtual image and upload/process some number of samples (we recommend at least 200GB to start). The virtual image contains all of the analysis functionality in a single panel and does not require any additional software other than the previously mentioned virtualization host.
For FusionPlex® RNA Assays: A sample will pass the QC filter if the average number of unique RNA start sites per control GSP2 (Gene Specific Primer 2) is 10. This metric is reported under the Read Statistics tab, labeled Average Unique Start Sites per GSP2 Control.
For VariantPlex® DNA Assays: A sample will pass the QC filter if the average number of unique DNA start sites per target GSP2 is 50 or above. This metric is reported under the Read Statistics tab, labeled Average Unique Start Sites per GSP2.
This has proven to be a reasonable criterion for the success of the assay. However, users are encouraged to evaluate the results themselves and set their own cutoff.
For installations requiring only a single sample to be processed at a time, we have extensively tested Oracle VirtualBox (for workstations). Oracle VirtualBox can be downloaded from virtualbox.org. However, because VirtualBox does not properly support multiple processors, we require VMWare on any installation where more than 1 sample needs to be processed at a time.
For single workstation solutions requiring multiple sample processing, VMware Workstation Pro is recommended for Windows, and VMware Fusion is recommended for Mac. However these VMWare products support a maximum of 64GB RAM to be allocated.
For installations requiring more than 5 samples to be run concurrently, we recommend VMware EXSi.
VMware products can be downloaded from vmware.com.
Archer® has provided information to guide the detection of fusions and variations. However, every sample is different. Each lab must interpret the results, and take into consideration any information known about their particular samples. Read statistics and the visualization tools can assist in making proper calls. Some things to consider include:
- Does the fusion appear in the strong evidence tab?
- Are there several reads supporting the fusion call? 5 is the MINIMUM recommended number of unique reads to call a fusion.
- Are there enough reads supporting the fusion transcript(s) as compared to the wild-type transcript(s)?
- Is the expression level of the target gene normally high or low?
- How does the expression level of the target in the analyzed sample compare to expression in a normal (control) sample?
- Is the fusion known in the literature?
- When visualizing the data, are there any insertions or deletions present in more than a few of the reads near the breakpoint? This may indicate a mapping error.
Watch the video below to learn more about visualizing reads and fusions in JBrowse:
If a fusion is in the Archer® Quiver® known fusion database, we always display it in the Strong Evidence tab.
In all other de-novo cases, the read event must meet the following criteria:
- Have a minimum of 5 unique (de-duplicated), breakpoint-spanning reads that support the gene fusion.
- Must not be flagged as a likely run cross-contamination event.
- At least 10% of the total reads for that GSP2 must support the fusion over the wild-type transcript.
- Have at least 3 unique start sites; breakpoint-spanning reads that support the gene fusion.
- Be an exon-exon fusion. Intron-exon fusions will be displayed in the weak evidence bin by default.
- Fusion partners cannot have homologous sequences. Fusion partners with too high sequence similarity will be flagged as likely mispriming events.
- Fusion partner genes must not be known paralogs.
- Fusion must pass additional alignment quality criteria (BLAST scores, mismatches in alignments, number of fusion partners involved, number of bases aligned to each fusion partner) to appear in the Strong Evidence bin.
These are the default values, however users can adjust these values in the Analysis User Settings based on their own data.
This reduces the total number of fragments that are analyzed and may cause the sample to fail QC. Poor quality samples have lower percentages of unique fragments. Make sure you are following the extraction recommendations.
Archer® Analysis utilizes BWA and Bowtie 2 for mapping to hg19. Any read that has an Alignment Score less than 30 is removed from consideration. Reads with a mapping quality less than 30 are either multiple mapping reads or contain many low-quality bases (low FASTQ base quality). For the exact calculation of the mapping quality and the factors involved, please refer to the BWA and Bowtie 2 documentation. Note that this default value can be adjusted by the user under Analysis Settings in Archer Analysis.
Variant reporting is even more powerful in Analysis 4.0. You can now apply specific variant filters, customize how and what information is displayed, and save filter sets for repeated use.
From the sample summary screen, click “Detailed summary” below the sample name. Then, click on “Variant Summary”. From here you can control your filter sets, add filters, hide or display columns, or export your data.
First, create a new filter set by clicking the dropdown and selecting “new”. We’ll call this one “somatic with targets”. After this filter set is saved, you can start adding some filters. For certain filters, such as this consequence filter, you can select multiple values by holding down the command key on a mac or the control key on windows while clicking. You can create as many filters as you want, and when you’re done, hit save.
Next, click “edit columns” to change the visibility and sorting of each column. You can also view information about that specific field, and when you’re done, simply click “save”.
Archer Analysis now makes it easier to view VEP and VCP annotations within variants. Simply hover over the Annotation icon, and Analysis displays the related annotations. You can even sort based on whether annotations are present or not, and set the default annotation for each variant call.
After you’ve organized your data the way you want, you can choose to export only your filtered data in TSV format or download the source file containing all of your variants in VCF or TSV format. This way, you can view your variants in the external application of your choice.
With Archer Analysis, it’s never been easier to organize variants and take control of your data. Learn more about Archer Analysis and test-drive it today at Archer Analysis.
There are three ways you can access Archer Analysis:
- You can use our live demo site at: analysis.ArcherDX.com. However, this is meant for training and demonstration, and is not for analysis of confidential samples. It is also used by other customers and staff and may become very backed up.
- The recommended method is to download the Archer Analysis virtual machine. You will need to install virtualization software (VMware) and download our .ovf file, which can be imported by the virtualization software.
- You can purchase Analysis Unlimited, which is a private cloud-based instance of Archer® Analysis where you pay only for the data you use. Contact technical support at tech@ArcherDX.com for more details.
Contact email@example.com for further information and to request the user guide.
This functionality is enabled through a mechanism called “Watched Folders”. A user with the appropriate permissions may configure a directory on the server for Archer Analysis to monitor. When sample files or a folder of sample files appear in the watched directory, Analysis will automatically process these files according to the Watched Folder template.
Typically a demultiplexer would be responsible for performing read assignment and copying sample files to these watched folder locations. This is beyond the scope of Archer Analysis itself but we can provide assistance in helping you configure this in some scenarios.
Please see the user manual for instructions on setting up this feature. For a copy of the Archer Analysis User Manual contact firstname.lastname@example.org.