The ClinicalGenomicStatement (CGS) template is the main vehicle of representing structured data in the GTR Implementation Guide. At its core, there is a genetic observation finding (result), e.g., a genetic variation, which can be associated with the following: more observations that describe the core observation in more detail (GenomicAssociatedObservation), e.g., amino acid change type; indications for performing this genetic observation (IndicationObservation), e.g., the genetic disease being assessed; and interpretations of the observation finding (InterpretivePhenotype), e.g., the mutation is pathogenic. In addition, it is important to associate a specimen and genomic source class with the core genetic finding, and optionally other performers of the genetic observation or the interpretation than those listed in the GTR header could be specified as part the CGS template.
The overall structure of CGS is depicted in chart 5 of the slide deck enclosed in the GTR package.

Due to the complexity of the interpretation of genetic observations, this template disallows the use of the interpretationCode attribute, rather uses an association to InterpretivePhenotype. The latter is constrained by several of its sub-templates in order to suggest binding to LOINC value sets describing interpretations of genetic testing. For example, ClinicalGenomicStatementGeneticVariation is a sub-template of ClinicalGenomicStatement and is associated with InterpretivePhenotypeGeneticVariation which is a sub-template of InterpretivePhenotype and its code and value attributes may be bound to LOINC codes representing interpretations of genetic findings.

All GTR structured data entries shall be part of CGS instances so that parsing applications can find the full semantics explicitly represented in this single and coherent structure. Sub-sections such as Indications, Interpretations and Specimen are mainly for presenting the data for human viewing (narrative), but they may also contain structured data that shall reference or be referenced from CGS. By using the referencing mechanism (explained in the GenomicObservationReference template), it is possible to have less redundant documents, e.g., in the case where all tests reported in a GTR document have the same indication, an Indications section in the Summary section consists of a full-blown indication observation which all CGS indication observations reference.

Note that CGS structured entries may point to the respective narrative in sub-sections by means of XML ID (see examples in the GTR samples enclosed in the GTR package). This mechanism can be useful in cases where the structured data has been generated by automated processes rather than by human coders.