I have had the privilege of attending this conference every year (except for one) for over 10 years, and this year has a few memorable experiences, presentations, conversations and introductions. Companies come and go; the mix of attendees is of course different every year; and I’ve had enough change in my own professional affiliation that often the first thing people do when they see me is to note what company is listed on my badge!
Here are a few highlights that give me some things to think through as a result.
The capability to do long-range phasing was what 10X Genomics launched with 3 years ago at #AGBT15, and with the advent of scRNA-Seq the progress has been palpable. Something like 25% of the talks were around single-cell genomics, and to have a packed audience hear about CBGB’s for CNV detection from single cells (awesome shirts, by the way), the ability to do sc-ATAC-Seq as an end-to-end application, and the ability to have 81 antibody labeling of single cells as a third application is a clear demonstration of the power of this technology.
That last point, on the antibody labeling, was lost (to me at least) amid all the other news. Called CITE-Seq in one paper (September 2017 Nature Methods), the idea is to use simple strepavidin-labeled antibodies to biotinylated barcodes that get sequenced and read out along with the barcoded transcriptome data.
It was a pleasant surprise to hear from Dr. Oppert from the USDA ARS with the title “A tale of three beetles: Lessons learned from sequencing the genome of stored product insects”, and using high-coverage Pacific Biosciences long reads for their genome assembly. Even with the latest iteration of the CANU assembler, it took a technology such as Hi-C from Dovetail Genomics to increase their scaffold length substantially.
While difficult to capture all the tables of data, it was memorable scanning to the right column of the summary statistics to see that the longest scaffold would increase some 9x or 30x or even higher. They are able to assemble entire chromosomes, this is a great technical advance.
And then the note about food security, and how dependent we (as Americans) are dependent upon only 12 plant and 5 animal species for survival, with all of its attendant problems (such as consumption of water and problems with CO2 waste). Why not rethink our diet, and consider the lowly cricket. It gave me reason to think that vertical cricket farming should be in my future. Or something like that.
And then there was the wonderful talk by Yaniv Erlich on DNA Media. What was there not to like? Afterwards I did feel like I was back in 1956, with only that gigantic 305 RAMAC that leased for some $3,200 per month for 5MB of 24” magnetic disk storage. It is too hard for me to think that reading and writing with the existing equipment and technology can progress, and I am firmly in the equivalent of 1956. Imaging a day where 128GB is in my hand (that is 25 of these RAMAC units that each weighed over 2,000 lbs and had to be delivered by airplane) is just too much of the strain on the imagination.
Later while talking with friends I realized that what Yaniv described with this paper calling it DNA Fountain can be likened to the old USENET Parchive (parity archive) where missing files can be recreated. Yes, there is a reason why we have get-togethers after the long AGBT days (breakfast at 7:30am, vendor talks at 8am, plenary talks at 9am, evening talks end at 9:30pm, and social events end whenever people want to end). We talk about Parchives and .par files recreating USENET missing files, and someday DNA Fountain is going to recreate information synthesized by a yet-to-be-created technology that’s optimized for DNA storage of information.
For what it’s worth, I heard about a party that went until 4am today. No I wasn’t among them, I’m a little too old for that. But there was that one get-together at the Marriott in 2013 that I’ll never forget, chatting away until 3am.
Yaniv has kindly shared his slides here if you’d like to get a flavor of what he talked about. In the meantime I’m going to go back to my 1956 thinking.
One last word
Chris Mason of Cornell-Weill Medical College has to be one of the most prolific people around. He field-tested the Illumina iSeq, he did an infectious disease experiment on NanoString upcoming single molecule Hyb & Seq, he talked about the Space Twin Study on two occasions and beautifully stitched them in such a way that both people who only caught one of the two talks or people who caught both talks could get fresh ideas and fresh perspective.
And there was that one slide where he must have been talking at about twice the normal rate, and according to Wikipedia is something about 175 words per minute for a newscaster. I’d put it about 250 words per minute for that one slide about all the kinds of tests they were doing for astronaut Kelly and his twin brother.
A friend Stephane Budel who is a principal of DeciBio (a life science consulting firm) wrote up his survey of 75 attendees. This occurred literally minutes after I finished writing this up, as I headed to an evening reception and end-of-conference party afterwards.
As usual his perspective on the NGS market is super-informative, and he even included a cool Tableau widget to slice through the market data. Find his write-up here.