AT A GLANCE: BSc in Pure Mathematics and Statistics from Coleg Prifysgol Cymru, Aberystwyth, Wales, in 1981. PhD in Mathematical Genetics at the Statistical Laboratory, Cambridge University, in 1985. Taught statistics at the Universities of Liverpool, Washington, and Bath before joining Myriad in 1995. Admits to being a “really bad guitarist and a really good fly tyer.”
QWhere will bioinformatics be in two years? Five years?
AMy only prediction is that in about five years a sequencing cult will develop. Although of little practical importance, followers of the cult will be obsessed with closing the final few remaining gaps in the human genome sequence. Occasionally some interesting new methodology will develop from this, but mostly it will be about printing up “I sequenced the chromosome 16 centrimeric gap” T-shirts and the like.
QWhat are the biggest challenges bioinformatics must overcome?
AFinding suitable staff. Bioinformatics requires serious mathematical or computational ability together with at least a strong interest, and preferably some education, in biology or genetics.
QWhat do you see as the next area for bioinformatics to address beyond genome sequencing?
AAssigning function to genes is the real problem, whether it’s done by analysis of the protein or by association with phenotypes. Most immediately I think that comparative analysis of genomes will be important.
QWhat hardware do you use?
ASun Enterprise servers, disk storage and workstations, PCs and Macs. QWhich databases do you use? AAll the usual public databases, our own proprietary databases, and some third-party databases within our pharmaceutical group.
QWhat bioinformatics software do you use?
AMost of the bioinformatics software we use was developed in house. This includes DNA sequencer signal processing, base calling, genotyping, assembly, quality control, data browsing and statistical analysis. This was mostly driven by necessity. The products were not available when we needed them, or were not suitable for high throughput and automation. There is more available today but the quality, for the most part, doesn’t currently match what we have.
QHow do you integrate your data?
AWe use web browsers as much as possible to access our various databases.
QWhose DNA chips do you use?
QHow large is your bioinformatics staff? Is the company hiring additional bioinformatics staff?
AWe have an informatics staff of 40+. This includes scientists, programmers, and systems administrators. Sequence and other data analysis happen within the scientific project groups. We are always looking for more people with the right skills.
QHow do you attract and retain qualified staff?
AWe try to make sure that they’re doing scientifically interesting work and using state-of-the-art programming tools and methods. The company’s extension of stock option bonuses to all employees also helps.
QHow is your bioinformatics unit organized within the framework of the company?
ABioinformatics is a group headed by a vice president reporting to the senior vice president of research.
QWhat projects are you working on now?
ALarge-scale mutation screening, genotyping, sequencing, proteomic and compound screening.
QWhat made you decide to become a bioinfor- maticist?
AIt mostly happened by accident. After getting my BSc in math and statistics I had to choose from a list of applied topics for my graduate course. Genetics looked as if it would be most fun and I was fortunate to get my first choice. Everything else seems to have followed from that.