NCI Biomedical Informatics Blog
- The Promise and the Challenge of Deep Learning in Pathology
- Predictive Modeling for Pre-clinical Drug Screening: Improving Models Derived From Observational Studies Using Machine Learning and Simulation
- Population Level Pilot: Population Information Integration, Analysis, and Modeling for Precision Surveillance
- Introducing the Data Commons Framework
- Modeling the Dynamics of Membrane-bound Mutant RAS to Accelerate Discovery of Novel Drug Targets
Community Code Resource Directory
This directory is intended to promote the exchange of community-developed digital capabilities supporting cancer research. The resources in this compilation are recommended by the community and will grow quickly to include capabilities across many biomedical domains. Check back frequently as new resources are added. To recommend a community-code resource for inclusion in this directory or to provide comments on this page, e-mail opensource [at] nci.nih.gov (Application Support).
Contributor: Daniel Rubin, Stanford
The AIM API allows users of commercial and open-source imaging workstations to collect, manage, and store structured information and to associate that information with imaging features in a platform-independent manner. It also enables researchers to collect case-report information in the context of radiology studies, and to integrate that data with images as the basis for analysis and computer-aided diagnosis (CAD).
Contributor: Yue Wang, Virginia Tech
BACOM is a statistically principled in-silico approach to accurately estimating genomic deletions and normal-tissue contamination in order to recover the true copy-number profile of cancer cells. The BACOM API is a cross-platform and open-source Java application that implements the whole pipeline of copy-number analysis of heterogeneous cancer tissues and other relevant processing steps. The developers also provide an R interface, bacomR, for running BACOM within the R environment, through which users can smoothly incorporate BACOM into their specific analyses.
Contributors: Margaret-Anne Story and Bo Fu, University of Victoria British Columbia, Canada
BioMixer supports ontology and ontology mapping visualizations in collaborative settings. In particular, BioMixer supports social interaction around the visualization. A user can send an existing visualization workspace to collaborators via email, as well as initiate discussions by adding notes to the visualization. BioMixer is accessible using a web browser and does not require the download or installation of any software. The tool supports the publication of visualizations by providing interactive visualizations that can be easily inserted into external websites. BioMixer supports users with diverse backgrounds and preferences by presenting multiple coordinated views that aim to engage the audience from different viewpoints.
Contributor: Matthew Eldridge, Cancer Research UK, Cambridge Research Institute
The caArray Importer extends the caArray Java API to provide programmatic access for the retrieval of imported array designs; creation, modification, and deletion of projects; uploading, validation, and importation of array data files; modification of file type; and file removal.
Contributor: Fred Loney, Oregon Health and Sciences University
The goal of caRuby is to reduce barriers to adopting cancer Biomedical Informatics Grid (caBIG) tools. caRuby simplifies interaction with caBIG application services by presenting a JRuby caBIG façade that supports the following: migration from legacy systems; incremental updating from source applications; extraction from a caBIG database; utility administrative tasks; workflow data transformations; lightweight web services; and site-specific user interfaces. There is a current implementation for caTissue.
Contributor: Lee Cooper and Carlos Moreno, Emory University
Copy-Number Modules is a collection of software modules developed by the Emory In Silico Research Center of Excellence (ISRCE) that defines a pipeline for calculating copy-number data for the REpository for Molecular BRAin Neoplasia DaTa (REMBRANDT) project. The data created by this pipeline have been used to perform a Genomic Identification of Significant Targets in Cancer (GISTIC) analysis to define significant copy-number alterations in the transcriptionally defined tumor subtypes identified by The Cancer Genome Atlas (TCGA). This pipeline operates on the raw Affymetrix 100K SNP CEL files and produces a list of altered regions for each sample.
Contributor: Yue Wang, Virginia Tech
DDN is an analytical tool for detecting and visualizing statistically significant topological changes in transcriptional networks representing two biological conditions. Developed under caBIG's In Silico Research Center of Excellence (ISRCE) Program, DDN enables differential network analysis and provides an alternative way for defining network biomarkers predictive of phenotypes. DDN also serves as a useful systems-biology tool for users across biomedical research communities to infer how genetic, epigenetic, or environmental variables affect biological networks and clinical phenotypes. Besides the standalone Java application, a Cytoscape plug-in, CytoDDN, has been developed to integrate network analysis and visualization.
Contributor: Kevin Jacobs, National Cancer Institute
Whole-genome association studies are generating unprecedented amounts of genotype data and require new and scalable computational approaches to address challenges involving storage, management, quality control, and genetic analysis. GLU is a framework and a software package that was designed around a set of novel conceptual approaches.
Contributor: LabKey Software Foundation
LabKey Server is open-source software that helps scientists organize, analyze, and share biomedical research data. It is a secure, web-based data-management platform that provides a flexible and scalable foundation for building applications customized to researchers' protocols, analysis tools, and data-sharing requirements.
Contributor: Jun Kong and Lee Cooper, Emory University
The Microscopy Image Segmentation tools, developed by the Emory In Silico Research Center of Excellence (ISRCE), support the segmentation of nuclei and angiogenesis regions in digitized pathology-slide images.
Contributor: Tahsin Kurc and Jun Kong, Emory University
Microscopy Image Segmentation — Grid Services developed by the Emory In Silico Research Center of Excellence (ISRCE) support the segmentation of nuclei and angiogenesis regions in digitized pathology slide images.
Contributor: Fusheng Wang, Emory University
A data model and a database, PAIS has been designed as part of Emory's In Silico Research Center of Excellence (ISRCE) project to address the data-management requirements of detailed characterization of micro-anatomic morphology through many interrelated analysis pipelines. The data model represents virtual slide-related image, annotation, markup, and feature information. This set of information includes a) metadata about images; b) context relating to specimens; c) human observations involving pathology classification and characteristics; d) algorithm- and human-described segmentations, features, and classifications; and e) a description of the computation being carried out as well as identification of input and output datasets.
Contributor: Yue Wang, Virginia Tech
PUG-SVM is an analytical tool for multiclass gene selection and classification. Developed under the In Silico Research Centers of Excellence (ISRCE) Program, PUG-SVM addresses the problem of imbalanced class separability, small sample size, and high gene space dimensionality where multiclass gene markers are defined by the union of one-versus-everyone phenotypic up-regulated genes, and used by a well-matched one-versus-rest support-vector machine. PUG-SVM provides a simpler yet more accurate strategy to identify statistically reproducible mechanistic marker genes for the characterization of heterogeneous diseases.
Contributor: Andrew Buckler, Buckler Biomedical Sciences
QI-Bench provides open-source informatics tooling to characterize the performance of quantitative medical imaging as needed to advance the field. These tools may be deployed internally within an organization or used for collaborative work across organizations. The data on which they work may be accessible only to identified individuals or more broadly in an open archive to suit the specific project purpose.
Contributor: Sharath Cholleti, Emory University
Region Classification Algorithms and Tools, developed by the Emory In Silico Research Center of Excellence, use a texton-based classification algorithm to classify a red-, green-, and blue-channel (RGB) image into tumor and normal regions.
Contributors: Sharath Cholleti, Emory University; Patrick Widener, Sandia National Lab
Region Classification Algorithms and Tools — Parallel Machines, developed by the Emory In Silico Research Center of Excellence, use a texton-based classification algorithm to classify an RGB image into tumor and normal regions.
Contributor: James McCusker, Yale University
swBIG is a web service that lets users treat caBIG data services as Linked Data. swBIG is available as a prototype representational state transfer (RESTful) service that converts requests for resources from linked data URIs into caGrid service calls to requisite grid endpoints. This service uses a representation of the NCI Thesaurus converted into a simple knowledge organization (SKOS) representation using web ontology language (OWL) to SKOS. This representation provides the ability to reason over concepts as instances in property value sets as well as in conceptual models.
Disclosure and Support
The U.S. Government, NIH, NCI CBIIT, and their employees and contractors do not make any warranty, express or implied, including warranties of fitness for a particular purpose, with respect to resources or tools listed in the Community Code Resource Directory. Links to other Internet sites are provided only for the convenience of users: NIH and NCI CBIIT are not responsible for the availability or content of external sites. See the complete disclosure statement.
NCI CBIIT does not provide support for the resources and tools listed in the Community Code Resource Directory. Users should contact the code submitter for support via the relevant link listed in the Directory. To recommend a community-code resource for inclusion in this directory or to provide comments on this page, e-mail opensource [at] nci.nih.gov (Application Support).