Speakers at ICG-13

Speakers at ICG-13




























Professor Sir Walter Bodmer has led the Cancer and Immunogenetics Laboratory at the WIMM since 1996 with major interests in (i) the fundamental genetics and biology of colorectal cancer and their potential applications, and ii) the characterization and population distribution of genetic diversity in human populations. His laboratory's major cancer interest is in the identification and functional analysis of cancer stem cells using colorectal cancer derived cell lines as a model, and using enriched cancer stem cells for preclinical investigation of novel drug responses. Sir Walter has a BA (1956) in mathematics and did his Ph.D.(1959) in population genetics under Sir Ronald Fisher at Cambridge University. He did post- doctoral work in molecular biology under Joshua Lederberg at Stanford and then was on the faculty of the Genetics Department in the Stanford University Medical School from 1962 until 1970, ending up as a Full Professor. He returned to England in 1970 to become Professor of Genetics at Oxford University, and then in 1979 became Director of Research, and later (1991) Director General, of the Imperial Cancer Research Fund. He was Principal of Hertford College Oxford from 1996-2005. Sir Walter became a Fellow of the Royal Society in 1974, a Foreign Member of the US National Academy of Sciences in 1981 and was Knighted in 1986 for his contributions to science. He was amongst the earliest to suggest the human genome project and was the second President of HUGO, the Human Genome Organisation. In 2013, he was awarded a Royal Medal from the Royal Society for seminal contributions to population genetics, gene mapping and understanding of familial genetic disease.


Approaches to complex trait analysis: the genetics of the human face

Walter Bodmer and Daniel Crouch

University of Oxford, Oxford UK

 Larger and larger bodies of data are being used for human genome wide association Studies (GWAS). This increases the risks of relevant stratification, and leads to the identification smaller and smaller effects, often with odds ratios no larger than 1.03. Such small effect variants, however significantly different from zero as quantified by the p-value or Bayesian equivalent, cannot be the basis for any improved understanding of the fundamental nature of the genetic control of such phenotypes, since it is simply not possible to assign such small effects to a complex trait in a meaningful way. The genes identified are in almost all cases ones where there is an a priori expectation of the possibility of their involvement in the complex trait being studied.

The one potential application in humans that does not depend on identifying the effects of individual variants is the polygenic risk score. This is a concept very similar to those developed originally for animal and plant breeding where continuous and more rapid improvement in, say, selecting for increased milk yield in cattle, can be achieved by genomic approaches that do not depend on knowing the precise function of each variant involved. The polygenic risk score only has direct value in the human context if the improvement in predicted risk of, say, cancer incidence, can lead to implementation of an appropriate procedure, in this case possibly more intensive early screening of individuals with a significantly increased risk of breast cancer.

Real advances in the genetic understanding of complex phenotypes will only come from the identification of specific variants with large enough effects to be able to study them and their functional consequences.

Our studies on the genetics of the human face provide an example of the approaches that are need for the careful definition of the phenotype in order to find such relatively large effect variants. To do this we have devised an approach to identifying facial features with high heritability, based on using twin data to estimate the additive genetic value of each point on a face, as provided by a 3D camera system. We have also used the ethnic difference between East Asian and European faces as a further source of face genetic variation. Principal components (PCs) analysis of the additive genetic values then provides a fine definition of the surface features of human faces.  The upper and lower 10% extremes of the most heritable PCs are used for looking for genetic associations.  Following this approach, we have identified three specific genetic variants with notable effects on facial profiles and eyes

Join Mailing List