 So, we will move on for the other Common Fund project that we want to update you about, and that's the Genotype Tissue Expression Project, otherwise known as G-TEX, and Jeff Strewing is the lead for that at NHGRI, so turn it over to Jeff. Great. Thank you. I'll be telling you a bit about the Genotype Tissue Expression Project, so called G-TEX. This is a Common Fund project led by NHGRI and NIMH, Eric and Tom are our co-chairs. These are some of the lead institutes and lead program people. NCI is also a major contributor to this at the sample sort of collection and front end of things, but we've fortunately got very active participation from many ICs. It's been a really fun project. The overall goal of G-TEX is to establish a resource database and associated tissue bank in which we can study the relationship between genetic variation and gene expression initially in largely reference or non-diseased human tissues. The sort of grander goal is to collect up to approximately 1,000 postmortem donors in which we collect multiple tissues from each and measure gene expression the donors will be genotyped. It's funded as a two-year pilot project, really just to see whether this idea is feasible at all. Can we get out there and collect these tissues and get good quality samples from them? So in this two-year feasibility project, the specific aims are to enroll 160 postmortem donors that will be unselected for disease status. They'll have a smattering of common chronic disease. And then we need to show we can obtain high-quality RNA from these multiple tissues. We aim to collect really about as many as we possibly can, as many as is feasible, as 30 or more per postmortem donor. We'll be getting a kind of parallel series of surgery donors where we'll have a handful sort of four to five tissues that are routinely discarded after surgery as a sort of zero postmortem interval kind of comparison group. And the pilot will allow us to calculate at least cis expression quantitative trait loci. This is the sort of overall kind of data flow with the autopsy donors on the left and surgery donors on the right. So our goal again is 160, it's relatively broad eligibility criteria. Donors can be aged 21 to 70 at the time of death. Largely unselected for disease status. They can't have disseminated cancer and a few infectious HIV-related things. But largely, we'll take most individuals. A blood sample will be drawn from each and we'll do the aluminum 5 million SNP chip on those and also attempt to establish a lymphoblastoid cell line. We'll be getting a skin sample that will undergo gene expression analysis, but we'll also attempt to establish fibroblastoids cell lines from them. And then collect the many other additional peripheral tissues. When we have permission for the brain, it will be sent in whole overnight ice to an NIH-supported brain bank. We'll get a small bit of cerebellum and cortex in the autopsy suite. The rest of the brain is then sent on for more specific dissection. One of the specimens will go for immediate histopathologic review. And this is a very important part of the whole project in that we'll have an expert pathology review and H&E slide available for every single specimen that undergoes gene expression analysis. All this is done at biospecimen source sites of the multiple aliquots that at the current moment are being fixed in packed gene tissue, preservative. One aliquot goes immediately to our laboratory data analysis and coordinating center, where they will extract the RNA and do all the laboratory analyses. If the RNA is of high enough quality, we'll do mostly RNA-seq-based gene expression analysis. Not incredibly deep, but we don't have unlimited funds. In fact, we don't have enough funds to do all the potentially sort of over 8,000 peripheral tissue samples that we hope to collect. We have a sort of wet biorepository which will coordinate the storage and histopathologic review. This is a very full slide, but these are sort of all the funded sort of groups. Again, all the tissue collection apparatus is coordinated and organized through NCI and SAIC Frederick through the Cancer Human Biobank through Carolyn Compton's Office of Biospecimen and Biorepository Research. We've made three awards for biospecimen source sites. One is NDRI out of Philadelphia. Most all the postmortem donors are coming through the National Organ Procurement Organization. These are the individuals who organize and identify people who can donate organs. So NDRI is working with both Philadelphia and Virginia Beach OPOs. All of our surgery donors, we have just one site. NDRI, they will come from Drexel University there in Philly or Albert Einstein College of Medicine. And we're funding an LC sub-study through NDRI as well to Laura Siminoff at Virginia Commonwealth University. Roswell Park Cancer Institute is a second biospecimen source site. They're working with their local OPO, the Upstate New York Transplant Service. And our third one is Science Care in Phoenix, Arizona. They're a whole body donation program that deals mostly in training, education, and research. They don't do any transplantation of tissues or organs, but they fall under many of the same sort of FDA and other kind of rules and regulations about screening donors. We have a wet biorepository at Van Andel Research Institute, and SAIC Frederick is kind of providing the kind of data coordination of the sample collection part of it. University of Miami has a supplement to receive the brains. They'll dissect out approximately 11 sub-regions that then go on to the Broad Institute, who is our laboratory data analysis and coordinating center, where really essentially all the molecular work, the gene expression analysis, and the basic CISQTL analyses will be coordinated. We've made five R01 awards for statistical methods development for this relatively unique data of having gene expression on multiple tissues from the same donor. The NCBI will provide both a sort of a database of the EQTL results from GTEX and other EQTL human studies, and that will be where the controlled access data is made available as well. A pretty complicated structure, all of this is done through contracts. These are grants here. As I mentioned, we have a relatively small LC sub-study. We're thrilled to have Laura Siminoff leading that. Their goal really is to sort of understand the factors that affect consent to tissue donation for GTEX so that we can make what we hope will be a scale up an even better situation and just improving the informed consent process in this postmortem setting all around. We've got a website on genome.gov that's more oriented to the public. Kathy Eng, the program analyst here at NHGRI, has been really helpful in getting that set up and getting that content in there. As I mentioned, one sort of important and relatively unique part of this is that every specimen in this that will undergo molecular analysis will also have an expert pathology review. This is just a picture of some skeletal muscle from one of the early cases. We did some of the early cases where we collect everything in four preservatives, including formalin, which is the typical preservative for clinical pathology labs in this pack gene tissue. The pathologists basically love this pack gene tissue and it really gives H&E images that are as good or better than the formalin. And in fact, for these early cases at least, we're going only with this fixative because it allows us to simplify the collection quite a bit. This is a very complicated collection in an autopsy suite where you have to get many, many hundreds of labeled vials with exactly the right tissue and exactly the right amount and in preservatives and lots of things recorded. This is Carolyn's group of pathologists who are reviewing all the slides. We made awards the end of last summer. End of August was our biospecimen source site awards. It took us roughly six or seven months to kind of get all the protocols in place and really sort of begin to get our feet wet from December through early in this year, we did what we called alpha experimental protocols again where we collected four aliquots into each of four different preservatives, the pack gene tissue RNA later, snapfrozen and formalin. Then we moved to the beta sort of protocol phase where they're relatively stable. We're still tweaking things a bit but everything is now going into the pack gene fixative. We had our first in-person meeting with our external scientific panel in the first week in June and we've been in the field for five or six weeks then and I think we had sort of three donors by then. There was a bit of concern then. Fortunately, our enrollment has gone up dramatically since then, it's really a credit to the biospecimen source site. These are incredibly professional groups that are accustomed to dealing with families at this time of crisis in approaching them for consent for tissue donation and organ donation and do some research protocols. This one's more elaborate and more extensive than what is typical but we've since sort of bent this curve. This is just sort of a blow up of these first few months but later in June then we had five donors. We had nine in July and we've had 15 in August. So we've sort of bent that curve and really I think many of us have breathed a big sigh of relief that the sort of donor enrollment has really sort of kind of taken off. I mentioned the donor characteristics of 21 to 70 and such they had to have a BMI less than 35. We let a pretty broad range that the autopsy had to start within 24 hours of death. The vast majority of these have been much earlier than that and in fact many of them are less than 12 hours or six hours or less. This shows some early and all this is really preliminary data. This shows some of the molecular sort of quality data. On the y-axis here is the RIN number, the RNA integrity number. This is a sort of surrogate for quality. This goes from zero to 10. The higher number the better. Generally things in the five, six or certainly seven in above range are gonna generally give you a very good gene expression results. These are the first I think 26 donors that met the eligibility criteria. So each of these points represents an individual organ and the RIN value for that organ. The symbols in sort of the blueish green, there are individuals who are solid organ donors. So these are individuals that generally would have been on life support on a ventilator sort of and declared brain dead or cardiac death there. They then provide an organ, a kidney or something for clinical transplantation once that is completely finished. So all the clinical tissue and organ donation proceeds before anything related to GTX happens. They tend to be lower post-mortem intervals but basically you're seeing that with the exception of a few donors where many of the tissues are not very high quality that we've really got the majority of donors with median RINs well above six which is our semi-arbitrary sort of cutoff. I think that the molecular quality is looking quite promising early on. This is essentially the same data just sliced in a slightly different way. Again, the RIN number on the y-axis here, now I'm showing the various tissues that we're collecting. You can see it's a very broad range of organs, pretty much all the internal organs and such. So each dot here now represents one individual for which tibial nerve was measured and this is the RIN value. Everything with the BB at the end of it are the tissues that are processed at our brain bank and then set on to the LDAC. Almost all of them are yielding very high quality RNA. Some of the organs that we'd certainly like to study or sort of have had some of the lower RIN values, particularly liver, that's the one where there's almost as much existing human data on sort of single tissue existing in the literature and publicly available. Those haven't been generally all that great but I think we're gonna have many, many organs representing a range of tissues and histology and things that are gonna give us really quite good values. The data analysis is the initial processing and everything is being done at the Broad, led by Gatti Getz, but David DeLuca has been very instrumental in coming up with a sort of customized fire hose analysis pipeline for the RNA-seq and integrating all this to get to the point to be able to do an NEQTL calculation. You can see our statistical methods, our one investigators is an incredibly strong group. They attended, many of them attended our last meeting as well. So we've been having nearly monthly teleconference calls where this group gets together and talks about which reference sequence we're mapping the RNA-seq to and just so that everybody's on the same page about how the data's being generated and they're then poised to make best use of this unique data. We've really only got quite preliminary gene expression data. Most of these samples have come in in July and August and it takes a couple of months to get completely through the RNA-seq pipeline. We took some of the samples from those early donors where we compared the various preservatives because there's not that much known about RNA-seq with RNA-later or PAX gene tissue, but using AFI expression arrays that the correlation was really quite good between the same organs done with different CDNA library, perhaps in different preservative methods and the RNA-seq is not completely through the analysis, but it's looking very good as well. The sort of initial quality metrics for the RNA-seq look very good for PAX gene tissue. The Broad has done a fair bit of number of experiments with non-G-tex tissue looking at different CDNA library preps because we're getting sort of between 100 to one gram aliquots. A relatively small amount of that goes into the initial RNA extraction. We'd like to do this on much smaller input amounts in this DSN light, this duplex-specific nucleus-light CDNA prep looks to be really quite promising over a range of RIN values of RNA quality values and with pretty small input amounts. We're, I think, going to start with about a half a microgram, 500 nanograms, but it looks pretty good even down to the sort of 100 to 10 nanogram kind of input amounts. We do have a fair number of tissues now of real G-tex tissues that over a range of RIN values to still kind of put all this through its paces to get through the RNA-seq pipeline and that's just not quite available yet. But I think sort of in general, the donor enrollment now has kind of taken off where we're quite pleased with that. We may be able to get more restrictive to get even shorter post-mortem intervals to take those donors that give us even the best quality. Those solid organ donors are giving sort of the best quality. Unfortunately, those are donors that have been on a ventilator and the brain is generally not appropriate for analysis in those individuals. So it's a bit of a trade-off. The other organs we can get very quickly but we generally aren't collecting the brains, the whole brains on those individuals. All this data being a common fund community resource will be made available to the scientific community. The sort of final policies and everything are still being worked out but we imagine quarterly DB gap releases. The individually identifiable data will be behind controlled access even though in the post-mortem setting, this legally wouldn't be classified as human subjects research. Every either donor, if it's a sort of pre-decedent sort of consent or their necks of kin provide consent or authorization for this, they know about this data that we're doing genetic research and making it available to the scientific community. So it'll be in a DB gap like controlled access process. We'll make as much of it open access as we can a lot of the results. Some of the sort of non variant containing gene expression values we expect to be able to make sort of freely publicly available. And that will be through this website which is up now and you can sort of browse some of the existing EQTL data sets. There isn't any GTX data there yet. We have a very strong external scientific panel. Ross is on this panel. He's our council representative. And many of them are a number of them. I think actually more than half we're able to attend the last meeting. And these are the principal investigators of the various funded sites. These are incredibly professional teams at the biospecimen source sites. I'm thrilled to be working with them. Harold Magazine is the PI at Science Care. John Lonsale leads a big team at NDRI. Barbara Foster at Roswell Park. Greg Korzanowski leads the SAIC Frederick team that's coordinating all of this. Kristin Nardley and Wendy Winkler are the PIs from the Broad. Scott Joule from Van Ander Research Institute and Deborah Mache. Sorry, I miss typed her last name there. It should be Capital M obviously. Is the PI at the University of Miami Brain Bank. And I'll end there and take questions. So this is a really exciting initiative. I'm just, are you collecting, maybe I missed it, are you collecting as well-paired samples where possible while the individual is still alive related to the issue of not so much the degradation of RNA and all of this, but the presumably very aberrant expression patterns of all sorts of proteins or genes and proteins. Effectively not. So we're collecting no pre-decedent tissues on the post-mortem donors. I guess the donors that would be solid organ donors, these are individuals that generally are in intensive care units on life support. And it's usually a several day process of going through the steps to establish that there's no, that there's brain death and such. So those would be donors, I guess, where conceivably we could, we honestly didn't go down that path just thinking that it was already complicated enough sort of getting permission to collect all these post-mortem. The surgery tissues are, is it kind of meant to be something of a surrogate of tissues that will be subject to surgical and anesthetic changes that might occur. But these, so these will be tissues that have no post-mortem interval, certainly. And so this will be skin and fat and muscle and artery and nerve that are sort of fixed within minutes to an hour or less after. So they're not from the same people, obviously. Sorry, I've gone on too long. Was that Carlos? So will T-shirt resources also be made available for say creation of cell lines? I mean, one of the big questions will be how different is expression there from EBV transform cell lines, which is where we know what we think we know about EQTL. That's right, we hope to be able to, I mean, we're attempting to establish those. So we're, so the blood sample will be the source for the genotyping. We're putting some of it in pax gene to get RNA expression, and we're gonna attempt to establish EBV lymphoblastoid cells, exactly for that reason. We're also getting a skin sample that's sent on in media. So these are sent, you know, it's like immediately. So they're, I mean, this is like, it's a symphony to package and send all of these things. They go to many different places all in different temperatures, different. It's an undertaking. The skin goes directly as well for fibroblast and potential IPS cell establishment. That was the next question. Those have been really successful so far. So I think of the first 13 donors, 12 of the 13, we've established a fibroblast cell line. One of them became contaminated and didn't transform. Of that same set of donors, I think only three or four of the EBV lines have transformed. The blood that's obtained post-mortem, now many of these were what I refer to as tissue donors. So these are people who have passed away or in a hospital, in a morgue maybe, or on their way to a morgue and people are approached, the families are approached to donate skin or cornea or things like that. So the blood is obtained from the subclavian or intracardiac. It's not incredibly cellular and it's sometimes several hours old. So we're hoping to get as many of them transformed as possible, but at least early on it's looking like that won't be quite as successful. And it'll be a resource that folks can, somebody could propose less genome sequence everybody, let's make IPSLs in three different ways. That's right, that's right. I mean, our external scientific panel told us sort of warn us that it would be, probably take us at least as long to figure out how to get the samples back out of the repositories and it's taken us to get them in. So I wouldn't, next year when I give an update, I'll be surprised if we have the processes completely in place to make them available. That's absolutely the intention or hope. Yes. I was curious at the high number that were tissue donors and just wondered about the, both pre and postmortem changes to the brain and that the majority of those with brain death have incredible brain swelling, no perfusion, et cetera. Are those samples usable? Generally no. So the people that are solid organ donors, the people who have been on a ventilator for usually more than a day before the cross clamping and the removal of the organs for transplantation, those brains we essentially are not sending on. We are taking, if we have permission, we're taking a portion of cortex and cerebellum, but though you're right. I mean, those, the brains are really not. Jeff, has there been any consideration of getting other members of the family, at least blood from other members of the families? It's been raised a time or two. We haven't gone down that path. Certainly something we could consider. I mean... It's always useful from a genetics point of view at least to have other family members. And, you know, obviously talking to these families is a very difficult and delicate issue, but if the family is together at that time and what you need is a blood sample, I don't think it would be that much, much more difficult to obtain. That would be certainly something to consider. I mean, one of the sort of blessings of it being a two-year pilot is that we're using it as that. We've learned a lot, obviously, just getting to this point. And so, you know, we'll have the opportunity to make alterations as we consider scale-up and hopefully adding many other kinds of molecular analyses to the existing samples and using the tissues for further, yes. Is there any restriction on the ethnic makeup or what is the current ethnic makeup of the samples that are coming in or how's that being dealt with? I don't have on the tip of my tongue the geographic sort of ancestral origin of the donors. We, there are no restrictions about racial or ethnic makeup of the donors. In fact, we chose the 160 in part, assuming that we would sort of take essentially all comers and that we would have enough of any one racial ethnic group to have enough polymorphic individuals. So it's a mixture. I know there are, I think, at least a handful of African-Americans in the first 30 or so donors, but it's essentially all comers. Great.