Alex Philippidis Senior News Editor Genetic Engineering & Biotechnology News
GeneTalk links rare-disease geneticists, stokes collaborations.
Distinguishing pathogenic gene variants from those without clinical significance is difficult enough to have brought a present-day meaning to the nearly six-century-old metaphor of finding a needle in a haystack. Over the past two years, however, researchers have found a few of these proverbial needles with the help of a two-year-old web crowdsourcing platform designed for investigators focused on rare diseases.
GeneTalk lets investigators filter sequence variants with genotype frequencies, inheritance models, and expert-curated gene panels. Researchers can annotate sequence variants, incorpoate knowledge from other GeneTalk users about specific mutations, and discuss the variants and their biological and medical implications with those other users.
Peter Krawitz, Ph.D., a researcher at the Institute for Medical Genetics and Human Genetics (IMG) at Berlin’s University Hospital Charité, co-founded GeneTalk in 2011 with Tom Kamphans, Ph.D., an IT freelancer specializing in algorithm design and programming.
Since then, GeneTalk has been used to identify diseases in at least eight new genes and, more importantly, has already begun to address a major difficulty in studying rare diseases—the building of patient cohorts essential for clinical trials of new treatments.
“You start setting up functional assays to prove that the mutation that you actually detected leads to the disease. But the first step toward building statistical proof is finding several unrelated patients. GeneTalk basically enables you to find more patients,” Dr. Krawitz told GEN. “Users can study a flock of genes and certain mutations, and tell the community, ‘Okay, I’ve got something here. I’ve got a patient here that looks this or that. And I’m looking for similar patients with mutations in the same gene. That’s basically how you could start building your cohort, or start a collaboration between geneticists, or clinical counselors.”
Finding Mutations for HPMRS
GeneTalk lets users find other investigators studying the same gene or variants, facilitating research collaborations. In one such collaboration, Dr. Krawitz and colleagues in Germany teamed up with researchers at Japan’s Osaka University to develop a diagnostic gene panel for targeting all known genes encoding proteins in the glycosylphosphatidylinositol (GPI)-anchor-synthesis pathway to screen individuals matching these features.
Researchers detected three missense mutations in the gene PGAP2 in two unrelated individuals with hyperphosphatasia with mental retardation syndrome (HPMRS). The investigators concluded that impairment of GPI-anchor remodeling also causes HPMRS, and that targeted sequencing of the genes encoding proteins in the GPI-anchor-synthesis pathway was an effective diagnostic approach for the subclass of congenital disorders of glycosylation.
“They basically showed that the mutation results in a glycosylation defect. This gave us functional proof that the mutation is actually disease causing,” Dr. Krawitz said.
Details of the research were published earlier this year in the American Journal of Human Genetics (AJHG). (See “PGAP2 Mutations, Affecting the GPI-Anchor-Synthesis Pathway, Cause Hyperphosphatasia with Mental Retardation Syndrome,” [Krawitz, et al., 2013])
Another example of how a GeneTalk user successfully used the platform to analyze data and launch a collaboration involved gene mutations linked to bone fragility.
Uwe Kornak, M.D., Ph.D., a research group leader at IMG, and Bernd Wollnik, M.D., of University Hospital Cologne’s Institute of Human Genetics, led researchers from Egypt, Germany, Spain, Turkey, and the U.S. in showing that hypofunctional alleles of the gene WNT1 cause a congenital disorder characterized by reduced bone mass and recurrent fractures, autosomal-recessive osteogenesis imperfecta.
“Our finding that homozygous and heterozygous variants in WNT1 predispose to low-bone-mass phenotypes might advance the development of more effective therapeutic strategies for congenital forms of bone fragility, as well as for common forms of age-related osteoporosis,” the researchers concluded in “Mutations in WNT1 Cause Different Forms of Bone Fragility,” published in AJHG.
Genetic Clue to Hand Malformation
Another GeneTalk user—Aleksander Jamsheer, M.D., Ph.D., of the Center for Medical Genetics at Poland’s Poznan University of Medical Sciences—used GeneTalk for analysis, then teamed up with Polish and German researchers that included Stefan Mundlos, head of both IMG and the development & disease research group at the Max Planck Institute for Molecular Genetics.
Through exome screening, the investigators identified the nonsense mutation R179X in exon 3 of the FGF16 gene of a 12-year-old Polish boy with metacarpal 4-5 fusion, a rare congenital malformation of the hand. The boy was negative for mutation in three other genes—NOG (Noggin), shown to influence the number of bones of the fingers; GDF5 (Growth differentiation factor 5), which has a role in skeletal and joint development; and the HOXD13 (homeobox D13) gene linked to limb patterning.
Sequencing of the FGF16 gene in a German boy with typical metacarpal 4-5 fusion revealed another nonsense mutation, S157X, according to the study, published in the Journal of Medical Genetics (See “Whole-exome sequencing identifies FGF16 nonsense mutations as the cause of X-linked recessive metacarpal 4/5 fusion,” [Jamsheer, et.al., 2013]). Researchers said their findings offered evidence for the involvement of FGF16 in the fine tuning of the human skeleton of the hand.
The network has grown since then to around 600 expert users, Dr. Krawitz added. Nearly all have joined over the past year, as GeneTalk started to gain visibility through booths at key conferences, and especially through word-of-mouth: “Now it’s a really gaining momentum. We have several new registries every day.”
Cancer and the Cloud
As GeneTalk grows, the network is looking to expand into other areas of study, such as cancer—where it hopes to offer solid assessments of variants and discussions about mutations that also play key roles as risk alleles for cancer predisposition syndrome.
GeneTalk is hosted on Illumina’s BaseSpace genomics analysis cloud computing platform, which directly integrates with the sequencing giant’s MiSeq and HiSeq instruments. BaseSpace automatically transfers data to the cloud in real time during a sequencing run. A 10-person lab, each user gets 1 terabyte of data free, with the second costing $250.
“For us, it’s a big advantage to address all the BaseSpace users’ need for a filtering tool. That’s a big opportunity,” Dr. Krawitz said. “There are really hundreds of users every hour analyzing data. For BaseSpace users, we hope to offer a good solution for doing the analysis.”
And like your smartphone, BaseSpace has its own growing number of apps. BaseSpace’s two dozen bioinformatics apps, developed by Illumina or third parties have been launched by individual users some 15,000 times. Apps include alignment, amplicon analysis, annotation of data, de novo assembly, library QC, metagenomics, resequencing, small RNA analysis, tertiary analysis, tumor profiling, variant calling, and visualization.
“We have about 12,000 overall BaseSpace users signed up. And we see around 2,000 people using it in any particular week,” Alex Dickinson, Illumina’s svp, cloud genomics, told GEN.
GeneTalk has its origins in Dr. Krawitz and colleagues at IMG being one of the first centers at Charité to apply next-generation sequencing in a clinical context. The researchers saw they could only solve cases for which they had access to an expert in their clinical discussion round— namely, a geneticist who could explain why a particular mutation was actually the one that caused a disease, and not others that were so rare that they could not be ruled out based on any database.
“From that experience, we concluded that you have to go global with your discussion round,” Dr. Krawitz recalled. “You need to be able to contact any geneticist in the world that’s an expert for a specific gene. So we wanted to make use of all the social web technology that’s out there, and to build such a network for our geneticists. That was the idea three years ago, and it got GeneTalk started.”