 We're going to hear three different parts of presentation and first what you're going to be hearing about, yes. So first what you're going to hear about for me on behalf of ENCODE team and the rest of the colleagues is a little bit of information about the ENCODE background. I'll tell you something about the approaches, the goals, and the achievements of ENCODE and I'll take brief questions there. And after that I'll move on to tell you about the functional genomics workshop that took place in March. You already heard that mention, Narek Green's director report. And then Elise Feingold will take over and do the heavy lifting and tell you about the functional genomics concepts that are coming up for counsel. So first I'd like to tell you a little bit about why we think this is important. Functional genomics is of course central to NHGRI goals. In part this is because non-coding DNA is very important for disease and also for gene regulation. So the vast majority of common disease findings and the vast majority of common disease heritability has been reported to lie outside of protein coding regions. Non-coding DNA variants are known to cause human diseases and alter human traits. For instance if we look at Fragile X syndrome, that's a Mendelian disorder where essentially all of the heritability is found as a non-coding mutation at a single gene. Or if we look at ALS, a neurological disorder, the vast majority, the heritability associated with the largest variant is a non-coding variant again, about 30% of the heritability with one variant. So in order to understand this, we don't have a regulatory code like we have for the genetic code, so we need functional genomics experiments to better understand the non-coding portion of the genome. So ENCODE is such a project run by NHGRI. ENCODE is the Encyclopedia of DNA Elements. And the twin goals of ENCODE, as you've heard, are to identify all of the candidate functional elements in the genome. That's of course an aspirational goal. And then also to share this catalog as a freely available resource. This is for anybody who has internet, no login, no fee for purchase. Taxpayers have already paid for this. And this is very useful for the study of the genetic basis of human disease. We also think that this is useful for people studying gene regulation. So quickly, the timeline of ENCODE, the project began in approximately 2003, starting off looking at 1% of the human genome. It's funny looking back now and thinking how bold and audacious that was to look at 1% of the genome. By 2007, the project was looking at essentially the entire human genome. And the sister project mod ENCODE was looking at the worm and fly genomes. By 2009, there was a mouse project. And the current round of ENCODE, which started in 2012, looks at the human and mouse genomes. So ENCODE does this by having been built upon decades of research into gene regulation. So many mechanistic studies over the years have learned biochemical signatures from events that appear to be causal for gene regulation. What ENCODE is doing in other projects and investigators as well is to reverse engineered to say, if we use these signatures, can we predict where genes may be transcripts from the genes in regulatory regions such as promoters, enhancers, and insulators? So I'd like to tell you some of the accomplishments of ENCODE. We've released thousands of data sets. And these are released pre-publication by the consortium. They're high quality data containing replicates. And the data are uniformly processed, the consortium data. We share software. We do this through the ENCODE portal and also through standard resources such as GitHub. We're working with partners at other projects to increase data interoperability. We're doing this with, for instance, Common Fund Roadmap Epigenomics and the International Human Epigenome Consortium. And what we're trying to do is standardize the ontologies that are being used across projects, the metadata that's being used, and even the data standards to make all of this data more useful to people outside of these projects. We've also developed an informed consent to allow unrestricted access, sharing of genomic data from participants. The main thing I'd like to emphasize to you is an accomplishment of ENCODE are the publications. As you heard from our director earlier, there are hundreds of publications from the consortium, but I'd like to echo his words. What's most important is that data are being widely used by the community. We've tracked at least 1,000 publications from the community using ENCODE data. And I'd emphasize these are not citations of ENCODE. That's a far larger number. These are publications that actually have ENCODE data in them. And they span a wide range of topics, human disease, basic biology, and methods development and software. Fully one-third of these publications are in human disease, emphasizing, I think, the translational value of the project. If we look at the human disease publications, they span, as you might expect or hope, a wide range of diseases that affect the human condition. Diseases that are funded by a range of institutes across the NIH, including at NHGRI itself. So does anybody have any questions here about ENCODE background before I move on to the workshop report? OK. So the workshop took place on March 10th and 11th of this year. We had a few thousand hits from different viewers as the workshop took place. I'd like to begin by acknowledging first people that did important work on this. Our workshop organizing committee, first and foremost. Eric Borowinkle, Carol Bolt, John Liss, and Aviv Regif did a fantastic job. I'd also like to thank my colleagues at NHGRI for working on this. The participants in this workshop span broad ranges of science, as I think you would like to see. We have people that develop technology and work on particular diseases. We have people that work in standalone labs and people that are members of lots of different consortia, including charge, GTECs, and IHEC. We had NHGRI council members and ENCODE external consultant panel members. We also had people that were funded from NHGRI, other NIH institutes, NIH intramural program, as well as a number of international funding agencies. So we think we had broad representation. And there were two main objectives in this workshop. First one is scientific, to discuss opportunities for understanding genome function using large-scale genomic studies. And the second, which we spent less time on, is an implementation topic. How could NHGRI structure topics to address these opportunities? So the workshop was video cast live. We had a few thousand viewers as the workshop was taking place. The materials are archived, so the agenda is available, the slides and the videos at the NHGRI website. And the workshop report will be coming soon. The workshop report that council members have today is a draft. It's pretty close to final, but the final report should be out soon. So I'd like to begin with the ending, the recommendations from the workshop. We heard that there was a strong need to continue genome-wide identification of functional elements. We heard that it was important to add in functional characterization, functional characterization efforts for these elements. We heard that it was important to begin to apply functional genomics assays directly to disease studies, something that had intentionally been out of scope within code in previous rounds. We also heard the need for increased community participation, primarily at the level of community-supplied samples for the consortium, also for the community to, also for the consortium to take in community data. So the workshop was structured around four basic types of topics. There was background information. We heard about mapping functional elements. We heard about disease studies. And we also heard about basic biological questions. For each of these, we heard some scientific presentations from members of the community, followed by discussions. And if we looked at the background presentation, some of the key messages that we heard are encoded as very useful for studies of disease and biology, as evidenced by hundreds of publications from the community using encode data. We heard that systematic mapping of data in diverse cell types continues to be important. We heard this is important beyond simply aggregating opportunity samples. We heard that it's important to profile samples that either are involved in disease or in the disease state. We heard that it'd be important to have better information about functional connections between regulatory elements and genes. Increasingly, people doing disease studies are becoming aware that a variant doesn't necessarily lie in the gene it's linked to. It may be one, two, three, four, five genes aware. Finally, we heard about the importance of a systems biology approach to integrate this knowledge. From there, we moved on to the mapping topic. And we heard again about the importance of unbiased mapping of elements. We heard again about the importance of 3D interaction maps. This is one technique that can be used to make predictions of which regulatory elements may be acting on which genes. We heard about the importance of single cell and small cell number assays for a variety of types of work. There was a plea for getting beyond mapping and understanding what elements do. This would help us to understand their causality and better understand whether they may or may not be participating in human traits and human diseases. Again, a plea for better community participation. And we heard a lot of interest in maintaining high data standards, maintaining efforts for interoperability and data visualization tools being very important. So we moved on to disease. We heard a lot of interest in getting a better understanding of the non-coding portion of the genome. A lot is known and a lot is being done about the protein coding portion of the genome already. We heard that it's important to both map and to functionally characterize genomic elements. And then we heard diverging views, two very different ideas on how best to do this. There was a lot of interest in prioritizing elements to characterize or in samples to characterize based on their known role in disease. There are also groups that strongly advocated we should do this by looking at samples and variants spread widely across biology. And the idea was brought up that, for instance, think about the lesson of what we learned from GWAS. We had just looked for variants in the places we thought to look. We wouldn't have found much of what it was that we found. There is call for better predictions of the effective genetic variation on gene expression. Typically today one finds a variant. It's not straightforward to say, what might this be doing in the expression of some nearby gene? Again, we heard about systems biology approaches, including a nice vignette about how systems biology would have made a line of research, much simpler had those findings been in place at the time the work was done. And we heard a plea for better sharing between the clinical and research enterprises. And we heard that this is primarily a matter of policy rather than technology, but it's holding back research in the area. Finally, we heard about specific biological problems. We heard about the value of systematic perturbations in order to learn about the connections between genes and also genes to regulatory elements. We heard that it's very important to maintain high data standards and that developing new data standards and pipelines for new assays would be an important thing to bring to the community. Some said that one of the most valuable things that ENCODE has done is develop standards and processing methods for what are now pretty standard data types. We heard again about it being important to characterize elements that have been mapped by ENCODE, that this would help us to better understand the ENCODE catalog that we already have. We heard again about the importance of bringing in community participation at the level of bringing in samples and existing data. We heard that what might be useful for some clinicians is having a gene centered view, being able to say, I think this gene is important in this disease or for this patient, what can I find out about ENCODE from this. Finally, we heard divergent views on the types of data that different users would need. We heard that some users find it very important for their work to have a large number of cell types that are profiled, even if it's only by one or two assays. Other users find it very important to have a large number of assays that are done, even if it's only in one or two cell types. There are also users that require a variety of assays across a variety of cell types. So we're trying to simultaneously support these different user communities. So in summary, where this left us as the highest priority is to expand beyond cataloging to understanding function of genomic elements. We think we could accomplish this by continued mapping efforts, a new functional characterization effort, increased community participation and adding and direct disease studies. And where this would help NHGRI and the community is that this would enhance genetic studies of human disease. So I'll stop there with the workshop and the background. Any questions before I hand this off to Elise?