Collaborations among clinicians, human geneticists and model organism researchers facilitate diagnosis and studies of undiagnosed conditions. Candidate causative genes and variants identified from a patient with an undiagnosed disease can be explored in a number of genetic model organisms. Using state-of-the-art genome engineering technologies in these model systems, one can assess whether the variants of interest lead to functional consequences in vivo, and obtain phenotypic information that may directly or indirectly relate to the patient’s condition. Integration of biological information from multiple species can be complementary or/and confirmatory.

The workflow of the UDN and MOSC. Patients with undiagnosed conditions apply to the UDN primarily through a website (UDN Gateway) that is hosted by the Coordinating Center. Application forms and past medical records are then screened by a case review committee to identify cases with objective findings. Once a patient is accepted, she/he will receive a clinical workup in one of the Clinical Sites. For most cases, WES or WGS are performed on the patient and immediate relatives by one of the two Sequencing Cores. In addition, untargeted metabolomics may be performed on patient samples by the Metabolomics Core. By combining the phenotype and genotype information, some cases can be solved without further investigation. If a diagnosis is not made, the clinical site submits candidate gene/variant information to the MOSC together with a brief description of the patient’s condition. The MOSC first performs a database search using the MARRVEL tool to aggregate existing information on the human gene/variant and its model organism orthologs. In addition, matchmaking with patients in other disease cohorts are attempted through collaborations to identify other individuals with similar genotype and phenotype. Once a variant is considered to be a high priority candidate, experiments to assess gene and variant function are designed by the MOSC investigators and pursued in the Drosophila Core or in the Zebrafish Core.

Strategy to “humanize” a Drosophila gene to assess functional consequences of a novel variant. (A) For most genes, the Drosophila Core of the MOSC performs functional studies of a patient variant by humanizing the orthologous gene in the fly. First, a fly gene that is most likely to be the ortholog of the human gene is identified using the MARRVEL tool. MARRVEL also provides a link to the FlyBase page that displays known biological function, transcriptomics and proteomics data, mutant phenotypes and available resources for the Drosophila gene of interest. If a coding intronic MiMIC is available, the Drosophila Core uses this as an entry point to study the gene. Through recombinase-mediated cassette exchange (RMCE), an artificial exon is integrated that functions as a gene trap, creating a strong LOF allele. This artificial exon contains a T2A ribosomal skipping sequence and a coding sequence for the GAL4 transcriptional activator (T2A-GAL4). (B) By crossing the T2A-GAL4 strain to a transgenic fly that carries a UAS-human cDNA construct (together with a deficiency of the locus or an independent mutant allele of the fly gene, data not shown), the fly gene can be humanized. When the gene of interest is transcribed, the splice acceptor (SA) in the artificial exon splices into the upstream exon. Since a transcription termination sequence (polyA) is present at the 3′ end of this artificial exon, the transcript is terminated, and the remaining portion of the fly gene is not transcribed. When this transcript is translated, a truncated protein that is usually nonfunctional is made together with a GAL4 protein. GAL4 is expressed in the same spatial and temporal pattern as the fly gene, allowing expression of the corresponding human cDNA under the control of the UAS element (GAL4 target sequence). By comparing the ability of the reference (wild type) and variant (mutant) to rescue the fly mutant phenotype, one can assess whether the variant of interest impacts protein function.

The workflow of the Canadian RDMM Network. RDMM connects Canada’s disease gene discovery projects with the Canadian model organism researchers. Investigators that work with yeast, C. elegans, Drosophila, zebrafish, or mouse are encouraged to join the network. Upon registration, the investigator provides a list of genes or genetic pathways in which they are experts. In parallel, a physician or a human geneticist submits a “connect application” for cases that they wish to find a model organism collaborator for. If the case is approved by the Clinical Advisory Committee, the Scientific Advisory Committee performs a search of the model organism registry and identifies an investigator that specializes in the orthologous gene. Upon matchmaking, the model organism investigator and the physician/human geneticist discuss a working plan and submit a proposal to the Scientific Advisory Committee. If the case is approved, funding is provided (1 year, Can$25,000) to generate a disease model and study the candidate gene/variant. The long-term goal of this process is to connect clinicians and basic researchers to establish a collaborative network across the country to facilitate rare disease research.