The TGMI Team – Grace Tiao

Grace TiaoEvery month, one member of the TGMI team will tell us why they are so committed to the vision of the TGMI, and share a bit more about their work and interests. This week we hear from TGMI team member Grace Tiao, who is a computational biologist at the Broad Institute.


What has been the main focus of your work to date?

For nearly five years I worked as a computational biologist in Gad Getz’s group at the Broad Institute to discover genes that harbor germline (inherited) mutations that predispose people to cancer. Last summer, I joined Daniel MacArthur’s laboratory to lead his methods development team, which focuses on designing the computational pipelines needed to handle and analyze enormous datasets of human genetic variation. Our group is best known for creating the Genome Aggregation Database (gnomAD), which currently contains 123,136 exome sequences and 15,496 whole-genome sequences, as well as the public browser where summary data is available to the public; but we also work on pipelines to curate data from rare disease patients and their families to help clinicians and researchers deliver diagnoses.


What are you most excited about in genetic medicine?

To practice genetic medicine, we have to have an underlying knowledge basis about the genetic origin and mechanism of the disease

I think what’s exciting about genetic medicine is that, almost by definition, to practice genetic medicine, we have to have an underlying knowledge basis about the genetic origin and mechanism of the disease. My sister is a clinician and one of the things I never realized before watching her progress through training and into practice is that much of medicine is a black box: physicians are permitted to prescribe medicines off label because many medicines are potentially efficacious for conditions outside the officially approved ones; and a surprising number of drugs are approved for medicinal use without scientists ever knowing what the mechanism of action is. The only things that have to be demonstrated are efficacy — that it “works” (has a measurable effect on the condition) — and safety. But that leaves an enormous gap in our understanding: the drug may be more or less effective under certain conditions or in certain people, and since we don’t know how it works, we can’t easily improve or tweak its applications to suit individuals.

Genetic medicine, by contrast, begins by attempting to ascertain the specific genetic source or cause of a condition, and based on that information, we can begin to unravel the mechanistic cause of a condition and potential treatments to address what’s wrong. It’s a much more intellectually satisfying and complete approach to discovery and treatment; and it also has the potential to provide diagnoses and treatments — or even cures — to people with truly unique conditions who have not been able to get them before.


What are you most concerned about in genetic medicine?

I’m concerned that the technologies being developed to deliver genetic medicine — CRISPR-Cas9 editing, for example — might be used to permanently alter the human germline, with unknowable consequences for individuals and populations. I think there might be a reasonable case to be made for altering the germline for heritable genetic conditions that are medically serious; and clearly a great deal depends on the specificity and efficacy of the technology used to edit human genetic material; but because the risks in general are simply unknowable, it remains a concerning proposition to me. Gene therapies, or other treatments that work on a somatic basis (i.e., on genetic material in cells other than the gametes), aren’t concerning for me in the same way.


Why did you get involved in TGMI?

As the senior staff member of the methods development team for gnomAD, I work on projects that would benefit from some of the specific objectives of the TGMI (e.g., developing a standardized framework for consistent variant and gene annotation). By the same token, there are areas where we can contribute to TGMI, because we’re already working on similar projects: for example, in building the computational infrastructure to deliver fast, automated, and high-throughput data to researchers around the globe. Pooling intellectual resources and manpower to achieve these goals together makes a great deal of sense!


What is the most important thing that you would like the TGMI to achieve?

To save lives, and to improve the quality of the lives of patients through the tools and processes we develop for genomic research.


If you had a magic wand (i.e. unlimited people/resources) what would you do to make genetic medicine work?

I would have a magic decoder that could translate genetic variants into their functional consequences — particularly in non-coding regions of the genome. Interpreting variants remains a major challenge for the field and is crucial to our ability to diagnose and treat genetic diseases.


Do you have a favourite gene? If so – what and why?

Given that most of the genes I have studied to date are disease-associated genes, it’s hard for me to think of any of them in the positive sense of a “favorite”… But I will  say that one of my favorite recent research stories was the discovery from Bradley Bernstein’s group that disruption or hypermethylation of a CTCF binding site near PDGFRA, an oncogene in gliomas, causes a constitutive enhancer in a neighboring topological domain that is normally blocked from interacting with PDGFRA to come into contact with PDGFRA. The disruption of the CTCF insulator causes the enhancer to upregulate PDGFRA and contributes to oncogenesis. I think it’s fascinating that so much functional activity in the genome is confined to specific topological domains that are defined by these CTCF insulator proteins.

So much genetic research to date has been focused on understanding the effects of variation on first-level DNA structural organization — what we usually mean when we say we’re learning to “read” the genetic “code”; but DNA is a three-dimensional molecule with a three-dimensional topology that inherently constrains and structures its function. To lean a little more into the metaphor: learning to read means learning not only how letters are organized into functional units like words, but also the grammar of how those words interact, and how meaning is produced by a particular organization and collection of words. In genetics, we’ve learned most of the words and now need to learn the grammar of the genome. The existence of elements in the genome that act like punctuation marks — such as the sites where CTCF binds — is fascinating, and it calls attention to how much more there is to investigate in human genetics.


What is a surprising fact that few people know about you?

I have traveled to both the Arctic and Antarctic Circles — and unfortunately never caught a glimpse of either the aurora borealis or the aurora australis… but they’re both on my bucket list.


If you had a chance to experience a completely different career for a week, what job would you try?

Hands down, I would be an astronaut on a space mission: the International Space Station, the moon — I’d take whatever I could get!