Loading...
 

Dante Labs

Dante Labs became the newest entrant into the genetic genealogy test-only space. An Italian startup in 2018 with offices in New York City and lab services from BGI in Hong Kong, China. They are focused on a single Whole Genome product. They offer the FastQ, BAM and VCF file results of a 30x average read depth, 100 base-pair length read NGS test. Their coverage of the autosomes yields nearly 10,000x the SNPs of the traditional microarray approaches and is very comparable to FGC, ySeq and others with WGS products. Key difference being the cost. Dante Labs frequently undercuts the price by as much as 5x; offering flash sales for as low as $200.

Early results are indicate their yDNA coverage exceeds that of other Sequencing tests such as the BigY from FTDNA. They also provide the full coverage scan of the mtDNA in addition to the autosomes. Only 23andMe provides coverage of all your DNA in this manner utilizing a single test. They do have more of a focus at the medical market in mind with their developed reports delivered. They offer no match or even real genetic genealogy analysis services. One must take the data files provided and use 3rd party tools to process for submission to other sites such as GEDMatch, yFull, mitoYdna and others. But they are attempting collaborations with yFull and others.

Dante Labs is initially using BGI sequencing services in Hong Kong, China. At least for North America testing through 2019. European testing is claimed to happen in Europe with other equipment. BGI has a competitive product to Illumina, the dominant remaining supplier in the market (otherwise). BGI have consistently been undercutting the general market in cost. Dante Labs shocked the market when first offering their product for under $500. Then $395 on Amazon Prime Day in Jul 2018. And then yet again offering the product for $199 for a week in November, 2018 and again in November 2019 for the month. They have had other, similar flash sales for shorter periods.

In April 2019, they have come out with a new long-read sequencing product. Long-read technology is still being refined for accuracy and read depth. But this is a great advancement, when perfected, for getting better yDNA STR coverage. This test is being processed by their own equipment from Oxford Nanopore in their Italy lab / office. This service was removed a year later.

During the summer of 2019, Dante Labs setup their own sequencing center in Italy using three Illumina NovaSeq 6000 machines. Illumina made an equity investment that was paid in equipment and took a stake in the company. Starting with late summer, all new sequencing was done in Italy in weeks and the 6-18 month backlog of old, unfinished sequencing tests was (redone) in Italy over the next six months. For a few months, starting in December 2019, people regularly received results in 2-4 weeks after returning their sample. Then the pandemic hit and they went back into sporadic delivery for various reasons (shortage of library prep materials and flow cells, change in focus to Covid19 testing due to a large UK government contract, logistic problems with the shipping companies and suppliers, etc).

As of summer 2021, they appear to have done a reverse merger with Cambridge xxxxxxx. And are solidifying on different subsidiary company names for different services (KurixHealth for Covid19 testing, Stripe2be then Rare Diseases for the traditional WGS testing, etc. During this period of transition (of over 9 months and continuing), various websites have come and gone, various products and pricing have similarly come and gone, with many months there being no way for anyone other than USA-based customers able to order a new WGS test kit. The prices for the same product offered on the various websites ranged from $399 (seemingly to compete with Nebula) to $795. They had been settling on $599 as the standard, non-sale price before.

As with any short-read (150 bp or less) Sequencing service, many of the larger, longer STRs cannot be reliably extracted from the BAM result file. They often span the read length and so cannot be accurately counted. Or are too homogeneous a region to be reliably aligned to generate a count.