Name: Peter Pohl
Position: Co-founder and CEO, GATC Biotech, since 1999
Experience and Education:
Chair, board of directors, LifeCodexx, since 2009
Co-founder and CEO of LifeCodexx, 2008-2009
Diploma (MA) in public administration, University of Konstanz
Peter Pohl is at the helm of GATC Biotech, by its own proclamation the market leader for DNA sequencing services in Europe. Headquartered in Konstanz in the south of Germany, close to the Swiss border, GATC currently has 145 employees and 11 open positions.
The family-owned enterprise — Pohl and his two brothers own 98 percent of the firm's shares — has been in business for more than 20 years and has been profitable for all but one of these. Last year, the firm had about €13 million ($17.4 million) in revenues, and for each of the past few years, it has been growing by 20 percent.
GATC Biotech provides two types of sequencing services: custom sequencing of individual samples by Sanger sequencing and large-scale sequencing on next-gen platforms. Besides its headquarters, which currently provides both types of services, it maintains laboratories in London and Düsseldorf, which both offer Sanger sequencing. Next year, Sanger sequencing services will be consolidated in a single laboratory in Cologne, and the Konstanz location will focus on genome and diagnostic sequencing on high-throughput platforms.
The firm ordered its first 454 and Illumina sequencer in late 2006 and is currently equipped with three Illumina HiSeqs, one Pacific Biosciences PacBio RS, and two 454 GS FLX instruments. It also maintains about 880 CPUs and 280 terabytes of storage.
LifeCodexx, GATC Biotech's daughter company, is working on a non-invasive prenatal test for trisomy 21 that it plans to introduce in German-speaking countries next year. The firm recently forged a licensing agreement with Sequenom that gives it the rights to commercialize its test in Germany, Austria, Switzerland, and Liechtenstein. Sequenom has introduced a similar test in the US,
In Sequence visited GATC Biotech's Konstanz facility last week and asked Pohl about the firm's business, and the challenges of rapidly changing technology and mounting competition. Below is an edited version of the conversation.
Who are your customers?
You can divide our customers into two groups. One is academia: universities, customers like the Pasteur Institute, the Medical Research Council in the UK, the DKFZ [German Cancer Research Institute] in Germany. The other group is industry in several branches, for example pharma, chemistry, fine chemistry, agro companies, food companies, biotech companies in general.
What next-gen sequencing applications are in most demand?
They main applications asked for are whole-genome sequencing, not only of humans but also of bacteria, viruses, plants, and other organisms; targeted sequencing; and RNA sequencing. Those three are pretty much the same size.
How do you decide what sequencing platform to use for a specific job?
Since sequencing has been getting more and more popular, customers come to us who are not originally from the sequencing area. We ask them what they would like to do, and then we design the project with them, and we say, 'We think that for this request, the solution on this platform would be the best,' or we recommend combining two or three technologies to get the best result.
What kind of software do you use to analyze the data from the various platforms?
It makes sense to use specific standards that are established in the community. What we do is to build them into our own home-built pipeline, so the data can seamlessly walk through the software and come out [in a format] we want to deliver to the customer. So we look at what is on the market and combine that software with our own middleware into one pipeline.
What kinds of bioinformatics services do you offer to customers?
Some customers would only like to have the data, others really love to have a result and a report. It depends on the customer. We cannot do everything for everybody, but I think that for a specific request, we have a good portfolio to offer.
[For example,] we have a whole-genome pipeline for human genomes. We can use Eland [Illumina's aligner] in the pipeline, or other tools, but this is a specified pipeline that we set up, so we can deliver standardized results.
Who do you compete against these days? For example, there is growing competition from Complete Genomics, Illumina, and BGI.
We do face BGI as one of our main competitors. I think it's an unfair competition because they are highly subsidized by the Chinese government, acting as an academic platform but basically being a commercial provider, making contracts at subsidized fees. This is definitely a threat to our business, and also, in some way, to establishing scientific networks in Europe. Of course there are also other competitors, for example Macrogen in Korea.
You are a certified service provider for Illumina, but not currently a partner in the Illumina Genome Network, which provides human whole-genome sequencing services. How is that playing out for you?
We have been in close contact with Illumina right from the start of their sequencing business. When we ordered our first sequencer, it was still a Solexa, not an Illumina — Solexa was written on the machine. So we have a long history. Also with Roche we have a long history, working closely on upcoming technologies and projects. I think these are very good collaborations.
Nevertheless, sometimes it is irritating that some [contracts] for providing sequencing services are won by Illumina. As far as I understand, that's a business decision they have chosen to make, and so far, I cannot really complain about it; it's working out for both of us. For some things, we cannot compete with Illumina, and for some projects, we don't have the capacity — sequencing several thousand human genomes in a short time, for example — so for us it's fine that Illumina is doing that business. On the other hand, when we have the capacity, we say, 'Please acknowledge that this is unfair competition — it is not fair to compete against your own customer.'
Are you also competing against academic core facilities?
Again, that's an act of unfair competition. We were facing that also at the beginning of the last decade, when more and more core facilities started to offer Sanger sequencing, but now their number is decreasing. We are now facing, just in Germany, about 20 to 30 HiSeqs in core facilities, and they usually run at full capacity, which is of course a highly subsidized business.
I think this is, in many cases, not really efficiently done, and I don't think that money is well spent. I would say that it would make much more sense if science combines with the [commercial] ability to scale processes in an efficient way, but I guess that is probably a movement for the next two to three years. When more scientists understand that it's actually pure engineering they do, and not scientific work. Production of data is not what a scientist should do, they should interpret the data and gain knowledge from the data, and I think this is what their money should be spent on.
How do you deal with the rapid innovation of sequencing technology? How do you make sure you bring in new platforms early enough, but not too early?
We do very early technology scouting. Before we have acquired any new technology, we have met with the companies for several years. Usually, it takes the companies three to five years to come up with a new technology. We visit them at a very early stage, talk to them, see what their plans are for the next year. One year later, we come again, see what they have done in that year, and how realistic it seems to us that the technology will be developing [as promised].
The other thing is, do we think the technology has a future, just by its design? Talking about the next-next-next-generation, there are technologies coming up that will not use any chemistry anymore, which will be definitely a big advantage.
So the question is when to invest. Once you invest in a new technology, you have to keep in mind that these technologies usually don't work from day one but that they take some time until they are robust. Nevertheless, it makes sense for us to invest early because then we have the advantage of offering new services to early adopters among our clients, and see what the demands of the market are. So there is an investment phase in the beginning, but the idea then is that it pays back when we start offering services on a large scale.
What is the greatest challenge with bringing in a new sequencing platform?
Really, to have the right timing and to have dedicated people taking care of establishing the whole process because it is something no one has done before. These technologies are not like a smartphone, easy to use, so you need time to adjust to the technology and then make sure that you have robust processes in place. The other issue is that you need to have a broad customer base which you can address immediately with the opportunities of a new technology, and then really be able to convert that quickly into contracts.
You once brought in a new platform — Life Tech's SOLiD — and returned it to the company after a while. Did you bring that technology in too early?
With SOLiD, we did not see at that point what was the really big difference compared to Illumina, and then decided for Illumina.
You recently started offering services on the PacBio RS. How has that platform been received by customers? What are the most popular applications?
Definitely, the big [innovative] aspect of the PacBio is the long reads. Long reads are used for assembly, so that's one of the main applications at the moment. We see good interest in the PacBio, but nevertheless, it's in an early phase; we have to find out what the possibilities of the technology are, and how to use them best. So even though it has now been in our lab for two and a half months, it still takes time to adopt and to adjust. It does look fine at the moment, but there will be, as we were told, updates on the technology, and we are looking forward to those.
What new instrumentation are you planning to bring in next year, in particular for your planned diagnostic center?
We are looking into the new technologies, though I cannot say which ones we are looking at more closely. There are the obvious candidates, for example MiSeq and Ion Torrent. We will decide next year which ones we will establish in our lab.
Would you bring in one or the other, or could it be both?
It could be both, but adding new technologies adds new complexity, so one really wants to be sure that it fits into the rest and that it really offers stable quality, which in the end, the customer relies on.
The Joint Genome Institute recently said it's phasing out its 454's. Is GATC ready to part with its 454 sequencers yet?
Not yet. They are still doing some very nice work, also in the research diagnostic area. So at the moment, we don't see that we have to phase them out.
Can you talk about your expansion plans for next year?
We see that diagnostics is getting more and more interesting, and we are also seeing increased demand from industry, pharma, the agro and food business for studies and research programs. We will therefore concentrate on a genome and diagnostics center here in Konstanz, and expand that business, and move our Sanger sequencing business to a new lab in Cologne. We have seen that time gets more and more precious to our customers, and the logistics [for single sample sequencing] are much better [served there], and we can achieve shorter [turnaround times].
The genome and diagnostic center in Konstanz will have two different sequencing production lines, each with their own ISO-certified space. We expect to receive ISO certification probably in the first half of 2012.
So is GATC Biotech going to be offering diagnostic sequencing services?
At an early stage, we [already] do diagnostic studies that are still sort of in the research area. We do see that demand for those is increasing, and that more and more of our clients are requesting validated processes. Therefore, we are working on validated processes, getting certified, and then we can also offer these processes. At the beginning, we will be offering whole-genome sequencing, but in a second step, targeted sequencing will also make sense.
Do you see a trend towards clinical sequencing?
I'm not entirely sure whether it's [sequencing in] a clinical environment, but what we do see is that validated sequencing is what customers want to see and to have. We have run several pilot studies, and we have seen that there is an increase in demand. Not necessarily from hospitals, but a company or a university may also want to have validated results.
Generally speaking, how has the market for next-gen sequencing developed over the last few years?
We have been offering whole-genome sequencing for 20 years. In former times, it was done by Sanger sequencing, and at the very beginning, also with our own technology. What we then saw with the introduction of next-gen sequencing was a good opportunity, because with decreasing prices, there would be increasing demand for whole-genome sequencing, so we invested in those technologies. That was very wise in this case, because there would not have been any whole-genome sequencing business anymore within GATC otherwise. From 2007 to 2010, Sanger sequencing in the area of whole-genome sequencing completely decreased, and it's now been 100 percent substituted by next-gen sequencing. We still see an increase in demand, so we are growing in that area, and earning money with that, and that's how we want to continue.
Can you provide an update on GATC's daughter company, LifeCodexx? How did that firm come about?
I read in a magazine three years ago that a sequencing-based test for trisomy 21 was going to be introduced. That was a Sanger sequencing-based test, and I thought, 'Wouldn't that also be possible by next-gen [sequencing]?' We discussed that internally, and we found publications from Professor Dennis Lo, who is also involved with Sequenom, which we did not know at that time. We started an internal testing program for trisomy 21, and it worked out, and then we began a pilot study, which we recently completed, and it worked out, too.
Then we started discussions with Sequenom because they have a strong patent position and we wanted to combine our forces. That's why we asked for a license, which we recently obtained. Now, we are partnering with Sequenom, focusing on German-speaking countries. We are in the middle of a blinded study at the moment, results from which we expect in the first quarter of next year. And then, hopefully, we can start soon offering the test on the market.
Will you offer the exact same test as Sequenom, or will you have your own version?
We are in close contact with Sequenom, so we do exchange our experience, at least to the extent possible. To my knowledge, it is a very similar procedure, though there are still differences, for example in the evaluation of the data and the bio-IT pipeline that we have put in place. But on a broad spectrum, they are very similar.
Are you, like Sequenom, developing tests for other chromosomal abnormalities?
We are also developing tests for trisomy 13 and 18, and we have also started an internal R&D program for another test for preeclampsia.
Will insurance pay for the trisomy 21 test?
It will initially not be paid for by insurance but be an out-of-pocket test. Of course we would like to show that it's a very important test, and we hope that insurance is going to pay for the test in the future.
Where are you going roll out this test initially?
We already have collaboration contracts with the main prenatal centers in Germany. Also, the plan is that in the big cities in German-speaking countries, we will have partners who are interested in offering this test and will send samples to us. We will be doing the sequencing and deliver the information back to the doctors.
Any idea how much the test will cost?
We are discussing this at the moment. For amniocentesis, you are talking about prices of €800 to €1,200 [$1,070 to $1,600], and it will probably be in that range. Amniocentesis is paid for by insurance, in general.
What happened with your venture into the direct-to-consumer market?
That was an idea at the very beginning of LifeCodexx — at that time, called LifeCode — but when we started that, I have to admit that we were not aware that the actual information you could gain from such a [genotyping] test would be questionable. That's why we decided we don't want to do that. We were fascinated by the idea, I have to say. But looking at the results that you get, they are quite fuzzy, so we decided to stop that. Then we looked at an application where there is a medical need, and that's why we started our trisomy 21 testing.
A year ago, you announced that GATC Biotech was planning to sequence 100,000 human genomes by 2014. How close have you come to that goal?
We have analyzed about 500 human samples by next-gen sequencing since 2010. That sounds like a small number, but if one thinks back two years, this was a huge number, and I'm still positive on the 100,000 human genomes. The technology is developing so fast.