Bioinformatics Core

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

The Bioinformatics and Computational Biology Laboratory and Core at the NHLBI (https://bioinformatics.nhlbi.nih.gov/) utilizes a vast array of techniques and technologies from bioinformatics to statistics, as well as statistical and machine learning approaches to analyze and assimilate high throughput data and glean information pertinent to the underlying biology. We work to elucidate the links between mutational processes, genetic function, transcriptomics effect, epigenetics influence and phenotype by developing innovative analysis workflows and implementing relevant statistical analyses. In particular, we use cutting-edge computational techniques and methodologies for exploring and integrating massive Omics datasets from genomics, transcriptomics, epigenomics, proteomics, and metabolomics experiments, employ data science tools and skills, and develop Biological Knowledgebases to provide evidence-based information to clinicians, researchers, and ultimately patients. Rapid advances in high-throughput technologies have produced distinct biomedical data sets that can be analyzed using mathematical and statistical models including network science tools to decode interactions between functional molecules in living cells. Machine learning, on the other hand, can handle heterogeneous data in different ways such as naive Bayesian Network data integration, tree-based methods such as Random Forest, and penalized linear models that provide assorted integrative analysis of multiple omics data, by analyzing different omics layers together. In addition, the discipline of Network biology is rapidly emerging with most recent applications to personalized medicine. The driving goal of the Bioinformatics team at the NHLBI is to bridge the gap between precision medicine and emerging clinical and biological datasets, by developing biological knowledgebases and scalable interpretative tools that address the growing complexity of such data. Major objectives of the core are as follows: 1) Integrative Omics Data Analysis: including (but not limited to) Consultation and support, Project design, Grant support as well as analysis of Genomics (Genome, exome and targeted DNA-seq), Transcriptomics (bulk and single-cell RNA-seq, microarray), Epigenomics (methyl-seq, ChIP-seq, etc.), Proteomics, and Metabolomics. 2) Statistical and Machine Learning and Biomedical Data Science, to make sense of data, using statistical methods to explore association between variables, predict outcomes from features, and characterize similarity and discrepancy among samples via clustering. In the bioinformatics core, we help to develop new methods or to apply existing models/tools to address specific tasks including (but not limited to) hypothesis testing, classification, regression, and clustering. 3) Developing Applications and Biological databases: In house On-Premise as well as Cloud computing application development and automation workflow. Here at the NHLBI, like many large biomedical institutions, due to the increasing generation of genome data, there is the need to process large-scale genome data, store the data, efficiently share and retrieve data, all of which present major challenges. Therefore, we regularly develop in-house automation and cloud-based bioinformatics workflow platforms for data analyses, which enable reliable and highly scalable execution of analyses workflows in a fully automated manner. 4) Trainings, Seminars and Workshops (In-person training for staff and students, Seminars and Workshops): We have established and hosted several scientific seminars and workshops (https://www.lobos.nih.gov/bcb/training.shtml). These include an annual Workshop Series in Bioinformatics Analysis for Next-Generation Sequencing Data, a 13 week hands-on workshop for PIs, staff, and fellows, and an annual symposium on hot topics in the genomics field such as single-cell technology symposium in 2017, Epigenetics Sequencing Symposium in 2018, and Long-Read Sequencing Symposium in 2019. These symposiums featured presentations on basic and clinical research, computational data analysis, and emerging technologies in the epigenetics sequencing area. Various training sessions were organized with the intent of introducing our NIH research community to new tools and technologies and promote their implementation in ongoing/future translational and clinical research. In addition, in 2019, we also hosted a research topic at Frontiers in genetics ( http://bit.ly/39YXMtl ), entitled Machine Learning and Network-Driven Integrative Genomics with emphasis on topics like Machine Learning, statistical methods, and computational tools for building biological networks, and network-based analysis of disease. Scientific impact We routinely meet with PIs and/or researchers to discuss the types of analyses and support that our statistical and bioinformatics expertise can provide to help them design experiments. We help with grant proposals to incorporate statistical and computational biology methods or techniques. We also prepare letters of support for such grant proposals. The Core statistical and bioinformatics consulting and collaboration covers a broad range of study design, statistical analyses and data science issues including experimental design for preclinical investigations, population studies, and Big Data analytics, and statistical analysis of next generation sequence data and omics data. The core is also actively involved in statistical and bioinformatics methods research and development and collaborates with PIs, Postdocs, and Staff in writing and implementing competitive project proposals involving methodological and software development. In fiscal year 2020, the core actively contributed to 20 manuscripts and more than 35 posters and abstracts for conference presentation. More details can be viewed at https://www.lobos.nih.gov/bcb/publications.shtml