This article has been updated to include information about the Bench-to-Bassinet grant and to clarify details about software developed by Cincinnati.
NEW YORK (GenomeWeb) – The University of Cincinnati and Cincinnati Children's Hospital have in recent years launched a number of biomedical informatics initiatives designed to support large-scale, federally funded projects but also to encourage collaboration around pediatric diseases between clinicians and researchers within the institution and elsewhere.
Peter White directs the biomedical informatics division at Cincinnati Children's and chairs the biomedical informatics department at the University of Cincinnati's College of Medicine. He was recruited in 2013 from the Children's Hospital of Philadelphia where he helped established the Center for Biomedical Informatics in 2006. In his position at Cincinnati, he is responsible for, among other things, building out the institution's genomics and computational efforts, providing bioinformatics expertise and support services for investigators, and coordinating research and education efforts including developing a new doctoral program in biomedical informatics.
He is also leading efforts at Cincinnati to manage data for the Bench-to-Bassinet program, a multi-institution initiative launched by the National Institutes of Health's National Heart, Lung, and Blood Institute in 2009. The project focuses on the genetic and biological bases of heart defects in children. White worked on the project while he was at CHOP, where he was part of a team that built a data hub to manage and organize data for the project.
He continues to work on the project, now in its second iteration, at Cincinnati Children's. Earlier this year, the hospital received a five-year $5.7 million grant from NHLBI to manage data for the project. In total, the project has been awarded $32.5 million in grant funding over five years from the NHLBI and the National Institute of Child Health and Human Development. So far, the Bench-to-Bassinet consortium has recruited and phenotyped about 10,000 patients and some 11,000 parents at its nine member institution sites. Researchers have collected information on cardiac characteristics, family history, environmental variables, and more, White said.
A lot of the patients have also been sequenced or genotyped, he said. Specifically, some 2,000 probands and their parents have been genotyped. Another roughly 2,000 patients have undergone targeted sequencing using a custom panel of genes associated with cardiac defects. Also some 4,000 to 5,000 trios have undergone whole-exome sequencing while another 1,000 either have been selected for or are currently undergoing whole-genome sequencing, White said. Some patients have also had their RNA sequenced. So far, the project has generated about 150 terabytes of data. Some institutions are also creating animal models that have specific heart defects for the project, White said.
In this second round, the consortium will likely recruit a few more patients but it's not clear exactly how many, White said. Some of the newer patients recruited for the study will include patients that have both heart defects and neurodevelopment abnormalities. That's because researchers want to explore connections between neurodevelopmental problems and heart defects, White said. "There is a molecular connection there and some of the genes have been identified that may play a role ... so that will be a major focus of the consortium going forward."
Cincinnati also offers several computational tools and resources for Bench-to-Bassinet including pipelines for analyzing whole-exome, whole-genome, and single-cell RNA-sequencing data as well as for variant calling. They also have tools for integrating genomic and clinical data as well as for visualizing data, White said. The list of tools that consoritium members can use for projects includes ToppGene, which offers a one-stop shop for gene list enrichment analysis and for prioritizing genes based on functional annotations and protein interaction networks; and AltAnalyze, which is used to analyze splicing variants in microarray, RNA-sequencing, and metabolomics data.
They also offer HeartsMart, which provides Bench-to-Bassinet researchers with tools for integrating and querying clinical and genomic data. Researchers can use the tool, for example, to search for genes associated with specific cardiac defects that show up in patients of a certain demographic. HeartsMart is built on a software toolkit called Harvest that was developed at CHOP, and provides bioinformatics developers with reusable tools to build customized applications to explore and query information in large multivariate biomedical databases.
Meanwhile, Cincinnati is also involved in efforts to build a long-term repository of clinical records for infants diagnosed with genetic disorders through newborn screening tests. The resource helps researchers test new technologies and treatments as well as track health outcomes for children with rare genetic disorders. It includes tools for querying and exploring data and for identifying and exploring patterns in patients' clinical and molecular information.
The database, previously called the Long-Term Follow-Up Data Collection Tool, was made possible by a five-year, $1.8 million grant from the Newborn Screening Translational Research Network, an initiative created in 2008 with funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development. White led the effort while he was at CHOP and continues to lead the project at Cincinnati. In 2015, Cincinnati Children's received an additional $2.2 million to continue to develop and maintain the resource, which is now called the Longitudinal Pediatric Data Resource (LPDR). It currently holds data from 46 metabolic disorders and includes 1,800 common data elements and 9,000 disorder-specific elements, White said.
In addition to directing biomedical informatics at University of Cincinnati, White also co-directs the Cincinnati Children's' Center for Pediatric Genomics with John Harley, director of Cincinnati Children's Center for Autoimmune Genomics and Etiology. The hospital launched the center in 2015, he said, to support ongoing genomics activities and to organize resulting data more effectively.
One of its goals is to find ways to use internally available biomedical resources to improve care, management, and treatment of complex pediatric disorders. These include an 800,000 sample pediatric biobank and electronic medical records for patients going back 10 years. "We really want to be able to take our big buckets of data ... and be able to have those well annotated, managed, integrated and accessible to as many people as possible in our institution," he told GenomeWeb. "In addition, we wanted to galvanize all of the scientific activities that were already occurring around genomics [at Cincinnati] ... and to facilitate those going forward."
Another goal was to encourage clinicians and researchers to work on collaborative projects across many disease areas. To that end, one of the initiatives that the center launched about three years ago is a $5 million program that will support collaborative faculty pilot projects in the genomics space. The funds come from the hospital's research foundation. One of the requirements for getting pilot funding under the program is that projects have to include both clinicians and researchers. Besides funding, the center also works with teams to ensure that they have access to appropriate resources and expertise they need for their projects and that results are disseminated properly.
"This has allowed us to establish a really good culture across the institution ... where we have different researchers and clinicians talking to each other and engaged with each other," White said. "We've had virtually all of our clinical and academic divisions involved in these projects, many of them in multiple projects." The Center has funded about 21 projects to date in a number of disease areas including rare disorders. Most of these are discovery projects that use next-generation sequencing technologies but there are also several that use animal models or focus on patients.
As part of efforts to support genomics-based pediatric projects, the center is also building up its internal informatics and computational infrastructure. For example, last year, it began building a communal variant knowledgebase that will be available to all investigators at the institution who generate and work with genomic data. The idea is to ensure that researchers process, organize, and annotate data in a unified format. They will also share data from that knowledgebase with peer institutions who are interested in collaborations focused on specific diseases. "We are starting off with [whole exome] and whole genome but will quickly include other types of omic data," White said.
White also said that Cincinnati researchers are exploring avenues for collaboration with their peers at other institutions and is in the early stages of establishing a network of pediatric hospitals that share data across institutions. "A lot of what the Center for Pediatric Genomics and the [biomedical informatics] department are trying to do is to organize our data locally so that we can be an excellent participant in network-based research involving genomics," he said. "If our data is well organized ... then we can partner much more effectively with funders, philanthropists, industry, and other peer institutions."