We have developed an ontology to describe cancer cells and provide a framework for consistent annotation of cancer-associated cellular data. OncoCL captures the histopathologic and molecular properties of tumorigenesis at recognized stages in cancer initiation and at critical stages of cancer progression to bridge the gap between the languages used in conventional surgical pathology and in cancer molecular biology.
The conceptual basis of our approach is that cancer cell phenotypes derive from the succession of alterations based on acquisition of the cancer hallmarks described by Hanahan and Weinberg: self-sufficiency in growth signals, insensitivity to antigrowth signals, evasion of apoptosis, limitless replicative potential, sustained angiogenesis, tissue invasion and metastasis, as well as reprogramming of energy metabolism, evading immune destruction, genomic instability and mutation, and tumor-promoting inflammation. We have implemented a default logic representation based on linking to the existing Cell Ontology (CL) as a representation of the ‘cell of origin’ -- that is, a normal/canonical cell type that maintains all its normal characteristics until it undergoes oncogenic change. OncoCL extends CL to describe the molecular changes along the series of steps leading from a canonical cell, which is represented in CL, to a malignant cell. OncoCL makes use of a number of other mature biomedical and clinical ontologies to characterize the ‘mesophenotype’ of cancer cells by integrating genetic, genomic and phenotypic data related to the cell type using (among others) the Phenotypic Quality Ontology (PATO), the Gene Ontology (GO), and the anatomy ontology, UBERON.
This work is supported by NIH with funding through NCI CA155825.
OncoCL is free and open to all users. OncoCL related work has been presented at several conferences: