Understanding DISCOVAR output

A DISCOVAR de novo assembly is a graph. A typical assembly consists almost entirely of linear stretches, typically like this


which we call ‘lines’, and providing a rich data type that captures polymorphism and other important features. Further, with some loss of information, these lines may be ‘flattened’ into standard contigs. We have added a tutorial explaining how these data types are available as part of the DISCOVAR output. We are also interested in hearing your thoughts regarding the utility of these output types and others that might be useful to you.

Large genome de novo assembler released

We are pleased to announce the release of our new de novo assembler suitable for large genomes up to human size. This is an early release and should be considered experimental, but is fully functioning. Download it now.

Our new assembler, called DISCOVAR de novo (experimental), uses the same cheap data that the original DISCOVAR release does: 250 base paired-end PCR-free Illumina reads. No other libraries are required. The runtime for a human genome on a 48 core, 0.5 Tb server is only 36 hours, and produces an assembly with a contig N50 of ~100 kb.

We are actively developing DISCOVAR de novo, so check back often for updates.

Explore a de novo human assembly online now

Want a sneak preview of what we’ve been working on lately? Then check out this online demo that lets you explore a de novo human assembly produced by our new assembler DISCOVAR de novo.

Developed over the past 6 months, the new DISCOVAR de novo algorithm will be released later this summer. Unlike DISCOVAR, it can assemble large genomes de novo. It is also much faster, but still takes the same low-cost single-library input data that DISCOVAR does.

Whilst we prepare DISCOVAR de novo for release, take a look at the online demo we’ve set up. Here you can explore and visualize an assembly of the human cell line NA12878. You can enter any coordinates on the human reference sequence GRCh38, and the demo will show you the part of the assembly that aligns there. Using this tool, large structural variation events can be directly visualized, and simple SNPs appear as short bubbles.

Please check it out and let us know what you think via the forum.

HiSeq 2500 data quality

We have been asked what our DISCOVAR input data looks like, and the best way to answer this question is with some examples. We don’t claim that these data are necessarily representative, but they do illustrate what we are able to generate here at the Broad Institute.

VCF format now supported

DISCOVAR now generates variant lists using the Variant Calling Format (VCF). This is the standard used by the community and is supported by many tools. Whilst the VCF file contains all events found by DISCOVAR, the complementary .variant file may contain additional information not easily represented in the VCF format. We encourage our users to look at both. The VCF should be filtered prior to use, and we have provided a tool and instructions on how to do this.

To facilitate calling variants using DISCOVAR on large genomes, we have created a tool to merge VCF files generated for overlapping regions. Simply run DISCOVAR on each region in turn (or in parallel to speed things up), then merge the VCF files that are produced. We currently recommend using a 50 kb region size, with a 10 kb overlap.

For more information on the VCF output, filtering and merging, please refer to our manual.

DISCOVAR performance tips

Are you getting the most out of your hardware when running DISCOVAR?
Take a look at our Computational Performance tips – they could help you get more bang for your computational buck!

DISCOVAR is a heavily multithreaded and memory intensive tool that will push your machines hard. Configuring your hardware to get the best performance isn’t straightforward, but with the right settings you may see significant improvements. After much experimentation and investigation, and with help from fellow DISCOVAR users, we have prepared a set of tips. We’ll continue to add and update them as we learn more, and we would like to hear about your experiences via our forum.