The Role of Ontology in Big Cancer Data

From NCOR Wiki
Jump to: navigation, search

Date: May 12-13, 2015


Session 1: Addressing cancer big data challenges through imaging ontologies

Tuesday, May 12 in FISHERS LANE CONFERENCE CENTER, 5635 Fisher Lane, Rockville, MD 20852

10:00 Warren Kibbe (NCI): Introductory Remarks

10:10 Barry Smith (Buffalo): The cancer research ontology space: An introduction Slides

The goal of the meeting: to better understand the challenges involved in using big data for cancer research, and to explore the utility of ontologies in addressing these challenges.
An introduction to existing ontology resources in the cancer domain, including NCI Thesaurus and the OBO Foundry; addressing opportunities and reasons for scepticism as concerns the use of ontologies in addressing cancer big data; topics to address will include the ontology of tissues, of tissue banks, and of image banks
What are the desired outcomes from this meeting?

10:50 Ilya Goldberg (NIA): The role of imaging ontologies in cancer big data Slides

As image data accumulates descriptive metadata using controlled vocabularies (ontologies), two key applications form a key "Big Data" technology that will impact cancer research and medicine as a whole:
1. semantic search, which allows image retrieval independently of image content, for example, retrieval of images spanning different imaging modalities and scales.
2. automated annotation, which addresses the scaling problem in associating metadata with ever larger image collections by using ontological terms as categories and the images carrying them as sample data to train machine classifiers.

11:30 Break 11:45 John Tomaszewski (University at Buffalo) and Metin Gurcan (Ohio State): ​How ontologies can help in addressing the big data challenges of pathology imaging: perspectives from a collaborating pathologist and image scientist Tomaszewski Slides Gurcan Slides

Widespread availability of whole slide scanners has a transformative effect on pathology, resulting in data that is "big” not only in terms of size but also in richness of information. Computational algorithms have been developed to tap into this data both to assist pathologists and to improve diagnosis, prognosis and treatment. Now, however, there is an urgent need to provide a set of terms and formal definitions necessary to characterize both the histopathological images and the algorithms that operate on them. We will present in this light our on-going work to create an ontology for histopathological imaging.

1. Describe the transformative role of digital pathology

2. Explain the big data challenges in clinical research and computational algorithms

3. Outline how ontologies can provide solutions to address the big data challenges of pathology imaging

13:00 Lunch

Session 2: Addressing cancer big data challenges by capturing metadata for data integration and analysis through ontologies

13:45 Christopher R. Kinsinger (NCI): Ontology Considerations for Cancer Proteomics Slides

NCI’s Clinical Proteomic Tumor Analysis Consortium (CPTAC) conducts proteomic and genomic analysis on clinical tissue samples. While CPTAC has utilized controlled vocabularies for biospecimen resources, development and adoption of ontology tools for rapidly advancing technologies such as proteomics and genomics remains a challenge.

14:15 Anna Maria Masci (Duke): Understanding the tumor microenvironment: integration of multiscale cancer data Slides

Tumor biology can be fully understood by considering the tumor as part of an ecosystem that includes tumor cells, stromal cells, matrix, fluids and gas. Real Time PCR, Tissue Micro Arrays, Plasma data and ontology are approaches used to describe the complex tumor ecology.

14:45 Jingshan Huang (South Alabama): The OMIT ontology Slides

The purpose of constructing OMIT is to establish data exchange standards and common data elements in miR and microgenomics domain. Cell biologists and bioinformaticians can make use of OMIT to leverage emerging semantic technologies in knowledge acquisition and discovery for more effective identification of important roles performed by miRs in human diseases including cancer.

15:15 Break

15:30 Chris Stoeckert (Penn): Integration and alignment of ontologies for cancer metadata collection based on the Ontology for Biomedical Investigations (OBI) Slides

Cancer research is multidisciplinary and requires standard terminology from multiple domains. OBI and interoperable OBO Foundry ontologies can be used for collecting and integrating clinical and -omic metadata relevant to cancer data. The Ontology for Biobanking is an example based on OBI.

16:00 Gully Burns (ISI/ USC): Leveraging OBI to cancer pathways via Knowledge Engineering from Experimental Design (KEfED) Slides

Our biomedical knowledge about cancer pathways is derived piecemeal from small-scale experiments in studies of molecular biology. A single study may consist of as many as thirty individual experiments to test small aspects of the underlying hypothesis of a paper. Typically 'pathway databases' interpret and summarize this information as model constructs that describe signaling events with little or no reference to the originating data that acts as evidence for the pathway constructs. As a part of the RUBICON project within the 'Big Mechanisms' DARPA program, I describe our work to model of the underlying design of molecular biology experiments as part of a text mining approach under development across the Big Mechanisms program called 'reading against a model'. This application serves as a valuable method within which OBI terminology and structure is providing essential structure and support.

16:30 Close of Day 1

Session 3: Cancer big data and the Ontology of Disease: Addressing Cancer Big Data Challenges

Wednesday, May 13 in Balcony A, Natcher

Roles of Ontologies in Cancer Big Data

9:00 Larry Wright (NCI Enterprise Vocabulary Services): NCI Thesaurus and Enterprise Vocabulary Services: Resources for Cancer Research Slides

9:15 Lynn Schriml (University of Maryland Baltimore): The Human Disease Ontology: DO_cancer_slim - a unified representation of cancer disease terms Slides

9:30 Ada Hamosh (Hopkins): OMIM Slides

9:45 Olivier Bodenreider (NLM): Oncology in SNOMED CT Slides

10:00 Break

The Role of Ontology in Cancer Big Data Use Cases

10:15 Lindsay Cowell (UT Southwestern): HPV and cervical cancer data in Electronic Health Records -- A Big Data challenge Slides

what types of data do we need to represent? HPV and cervical cancer and the Infectious Disease Ontology; challenges involved in keeping and using large collections of samples and of sample data

10:30 Raja Mazumder (George Washington University): The need for cancer disease ontology for pan-cancer data integration and analysis. Slides

10:45 Susan Mockus (The Jackson Laboratory, Genomic Medicine): Using the Disease Ontology to Translate Pathology Reports to NGS Clinical Reports Slides

11:00 Peter Elkin (University at Buffalo): Ontology-based cancer biomarker discovery Slides

11:15 Discussion and Session Wrap Up

Ontology/vocabulary challenges, Use case challenges
Challenges, Needs, Action Item Solutions, short term steps, long term goals

12:00 Lunch

Public Session: Cancer Big Data to Knowledge

13:00 Mark Musen (Stanford / NCBO): CEDAR: Making it Easier to Use Ontologies to Author Clinical Metadata Slides

13:40 Judith Blake (GO / Jackson Lab): The Impact of Ontologies on Comparative Genomics for Cancer: The Human-Mouse Connection Slides

14:20 Cathy Wu (University of Delaware / PRO): Ontology and the Precision Medicine Initiative: The Role of the Protein Ontology in Cancer Knowledge Network Discovery Slides

14:50 Sherri de Coronado (NIH / NCI): The importance of Semantics in NCI Efforts to Achieve Precision Oncology Slides


  • National Cancer Institute Center for Biomedical Informatics and Information Technology (CBIIT)
  • National Center for Biomedical Ontology (NCBO)
  • National Center for Ontological Research (NCOR)
  • Center for Expanded Data Annotation and Retrieval (CEDAR)


  • Durga Addepalli (NIH / NCI)
  • Praveen Arany (NIH / NIDCR)
  • Nancy Beck (Reagan-Udall Foundation)
  • Judith Blake (GO / PRO / The Jackson Laboratory)
  • Olivier Bodenreider (NIH / NLM)
  • Evan Bolton (NIH / NLM / NCBI)
  • Jonathan Bona (PRO / Buffalo)
  • William Bone (NIH/NHGRI)
  • Gully Burns (Information Sciences Institute, University of Southern California)
  • Kisha Coa (NIH/NCI)
  • Sherri de Coronado (National Cancer Institute)
  • Lindsay Cowell (UT Southwestern Medical Center)
  • Rina Das ((NIH/NIMHD)
  • Valentina di Francesco (NIH/NHGRI)
  • Rao L. Divi (NCI Division of Cancer Control and Population Sciences)
  • Mary E. Dolan (The Jackson Laboratory)
  • Peter Elkin (Department of Biomedical Informatics, University at Buffalo)
  • Luis Espinoza (NIH/CSR)
  • Jianwen Fang (NIH/NCI)
  • Gilberto Fragoso (National Cancer Institute)
  • Gang Fu (NIH / NLM / NCBI)
  • Ilya Goldberg (Image Informatics and Computational Biology Unit, National Institute on Aging)
  • Sharmistha Ghosh-Janjigian (NIH/NCI)
  • Metin Gurcan (College of Medicine, Ohio State University)
  • Ada Hamosh (OMIM / Johns Hopkins University)
  • Lori Henderson (NIH/NCI)
  • Jingshan Huang (University of South Alabama / ITCR)
  • Rebecca Jacobson (University of Pittsburgh / ITCR)
  • Sonia B Jakowlew (BPRB / NCI)
  • Guoqian Jiang (Mayo Clinic)
  • Warren Kibbe (NIH / NCI / Disease Ontology)
  • Christopher ​Kinsinger (NIH / NCI)
  • Prasad Konka (NIH / NCI)
  • Raj Krishnaraju (NIH / CSR)
  • Leandro Hermida (NIH/NCI)
  • Jerry Li (NIH / NCI)
  • Lu, Le (NIH / CC /DRD)
  • Zhengwu Lu (NIH / NCI)
  • Diana Ma (Hippocampus Analytics)
  • Hala R. Makhlouf (Cancer Diagnosis Program / NCI)
  • Cheryl Marks (NIH/NCI)
  • Anna Maria Masci (Duke)
  • Raja Mazumder (Georgetown University / Protein Information Resource)
  • Elvira Mitraka (University of Maryland, Baltimore)
  • Susan Mockus (Jackson Laboratory for Genomic Medicine, Farmington, CT)
  • Mark Musen (Stanford / National Center for Biomedical Ontology and Center for Expanded Data Annotation and Retrieval)
  • Darren Natale (Georgetown University / Protein Ontology Consortium)
  • Natsuko ​Miura (NIH / NCI)
  • Lauren Neal (Booz Allen Hamilton)
  • April Oh (NIH/NCI)
  • Miguel Ossandon (NCI / NIH)
  • Lisa Paradis (NIH / NCI)
  • ​Praveen Arany (NIH / NIDCR)
  • Thomas Prince (NIH / NCI)
  • Dani Przychodzin (CSC / NIH)
  • Thomas C. Radman (National Institute on Drug Abuse/NIH)
  • Lyubov Remennik (NIH/CC/BTRIS)
  • Betsy Rolland (NIH/NCI)
  • Lynn Schriml (University of Maryland, Baltimore / Disease Ontology)
  • Payal Shah (C-Change)
  • Mukul Sherekar (NIH/NCI)
  • Hoo Chang Shin (NIH/CC/LDRR)
  • Barry Smith (Buffalo / Open Biomedical Ontologies Foundry)
  • Sriram Sridhar (Booz Allen Hamilton)
  • Chris Stoeckert (University of Pennsylvania)c
  • Shumei Sun (Virginia Commonwealth University)
  • Svetlana Radaeva (NIH/NIAAA)
  • Feng Tao (NIH/CSR)
  • Elizabeth Thompson (President and CEO, C-Change)
  • John Tomaszewski (Pathology and Anatomical Sciences, Buffalo)
  • Eric Weitz (NIH/NCBI)
  • Larry Wright (NIH/NCI)
  • Tsung-Jung Wu (NIH/NCI)
  • Cathy Wu (Delaware / Protein Ontology)
  • Tingfen Yan (NIH/NIMHD)
  • Hong Yu (University of Massachusetts, Worcester)
  • Shadia Zaman (FDA)