That’s the question Keith Grimaldi of the Eurogene project asks in the latest post on his newish blog. By “genetics” he means human genetics.
His answer?
Without genetics & nutrigenomics, epidemiological nutritional research will remain “mostly harmless”. Or to paraphrase a less amusing person maybe it’s like trying to govern the Italians — “not difficult, just a waste of time”
We’ve suggested something similar here a couple of times, albeit it much less eloquently than Dr Grimaldi. Are the people designing projects aiming to improve the nutritional status of communities, whether based on biofortification through genetic modification or diversity-based approaches, listening?
I think scientists at my institute, the International Livestock Research Institute, are asking a similar question regarding livestock genetic resources research—is it any use characterizing and conserving rare farm animals without placing this kind of work within a much broader genetics and genomics framework?
I agree, but I also think that these arguments sometimes suffer somewhat from the fallacy of hindsight. How do we know the future value of genetic diversity?
Biological conservation is different from keeping a collection of car parts with known functions, discarding obsolete items over time. Gene function may only be revealed in the future, in interaction with new environments. Evolution is fundamentally unpredictable. Also, environments change, new viruses evolve, and so on.
In the end, the broader framework, in which all this work needs to be placed, is evolution. But of course, a “broader genetics and genomics framework” should be part of this evolutionary perspective!
Thanks, Susan. Very interesting, and perhaps you can elaborate on that a bit further for us sometime? I was making a somewhat different point, but too telegraphically. That is, that how people make use of nutrients depends as much on their genes as on what they eat. You can’t really most effectively use plant diversity for nutrition without understanding human diversity.
Thanks for the comments and link Luigi – I have just posted a link to your blog in a comment to the post you mention
Keith
The genetic diversity of microbiota in our guts also seems important.
http://www.sciencemag.org/cgi/content/abstract/328/5975/228
And more.
Luigi: Let me elaborate a little here on what I meant by ILRI being interested in (increasingly) bigger genetics/genomics pictures/frameworks. (And if I get carried away to lay La-La-Land here, I request ILRI geneticist Steve Kemp to step in and save me.) ILRI has worked to characterize, conserve and better use the farm animals indigenous in poor countries for two decades. This requires going out and about to find out what is there, how people use their (often funky) native stock, what they value in them, etc. The scientists who typically do this kind of thing (formerly quaintly just called ‘animal scientists) have time to time needed ‘Indiana Jones’-type personalities as they trekked to inhospitable places to discover previously unknown breeds . . . And we still do this kind of thing. But some of our scientists (think lab coat rather then fedora) began to incorporate human genetics into their parasite and animal genetics work, which was a natural fit as many of our geneticists had medical research backgrounds and some of our target diseases (the protozoan sleeping sickness, for example) afflict people as well as animals. A few years ago, with the tremendous advances in gene technologies, a new kind of geneticist began to appear at ILRI, one that more resembles Hans Solo than Indiana Jones. These researchers are taking ‘whole systems’ approaches to the ‘livestock genetics’ field (a branch of knowledge that, like those temples Indiana Jones obsesses about, can appear of largely historical interest). These new geneticists are ambitiously adding environmental genetics (soil microbes, wildlife species . . . .) to their livestock, parasite and human targets of interest. They’re interested in the WHOLE picture—and they claim they have the tools to productively investigate this brave new world of ‘landscape genomics’. They also claim that their methods—which often now start with ‘big data’ rather than hypotheses—are ‘universally applicable for fast and productive collective mining by the livestock, crop and human health research communities alike’. (I’ll post a description of those methods in the next comment box.) So—we heartedly agree that ‘how people make use of nutrients depends as much on their genes as on what they eat’ and that to understand plant (and animal) diversity we must understand human diversity. But we are going much, much further these days, with some believing that ‘livestock genetics’ impacts may well be pretty ‘harmless’ themselves without making use of frameworks that take in (somehow, and ‘somehow’ I think is the catch) whole systems.
Herewith a description of the methods (by all those Hans Solo geneticists).
NEW SCIENCE IS NETWORKED SCIENCE
A global group of livestock geneticists and informaticians is using web workflows and networks to make their gene discovery tools universally applicable for fast and productive ‘collective mining’ by the livestock, crop and human health research communities alike.
Research methodologies developed at ILRI for identifying genes controlling disease tolerance in cattle are being adopted in many other fields. The genome data handling pipelines, data sets and bioinformatics resources have been developed as part of an ILRI-led Pathogen-Host research project funded by a Wellcome Trust to understand the nature of the response by cattle to infection with trypanosome parasites and the way this response differs between resistant and susceptible genotypes. The ILRI researchers employ mice as a model system, analyzing high-density maps of mouse genes, gene networks and sequence polymorphism to infer the location and function of cattle genes controlling bovine resistance to disease.
These methodologies are an intermediate output of ILRI’s Wellcome project to use functional genomics to identify a set of candidate genes controlling trypanotolerance. This work was published in two peer-reviewed papers and presented at three invited plenary presentations at major international meetings plus 13 other invited presentations. The key methodology paper, “A Systematic Strategy for Large-Scale Analysis of Genotype-Phenotype Correlations: Identification of candidate genes involved in African Trypanosomiasis”, was published in November 2007 in Nucleic Acid Research 35, 5625-5633.
This methodology paper describes a data-driven approach for locating genes involved in disease resistance. This approach complements the more traditional scientific one of setting hypotheses to test because it is (usefully) unfiltered by human assumptions.
The paper publishes the workflows used in the project. A workflow is a description of all the steps taken in a research study in a form that allows anyone to replicate exactly a complex set of queries across diverse datasets scattered around the world. Multiple workflows can be bolted together to allow new questions to be asked in a rigorous and repeatable manner. In this project, the workflows allowed systematic and large-scale analysis of microarray gene expression and quantitative trait loci, investigated at the level of biological pathways, which resulted in establishing links between genotype and phenotype
Informaticians in this research project addressed the big challenges the project presented in the capture, management and dissemination of data. They developed systems for rigorously describing data so that they may be used by anyone for anything. That requires that the data carry with them complete descriptions of both the experimental systems and the analyses performed. Importantly, this allows others to repeat the analyses with their own data, to perform their own analyses using ILRI data, or to “pull apart” the components of both data and methodology and then re-use the component parts for entirely novel purposes.
The methodologies this paper describes attracted much interest among scientists studying non-model organisms and among crop and medical as well as livestock researchers. The tools were included, for example, in a recent Nature review, “The New Networking [Science] Nexus” (21 Feb 2008), and are helping medical researchers understand the genetic basis of wound healing. The paper also triggered new clinical trials now being undertaken in the UK by intensive care specialists at the University of Manchester and Salford Hospital to understand the role of cholesterol as an indicator of the potential for recovery or death of critically ill patients under intensive care.
The workflows underlying the methods this paper describes are made available to the research community by means of a UK-funded ‘social networking for scientists’ website —
http://www.myexperiment.org/workflows — that lets users share customary protocols for standardizing data, running simulations and conducting statistical analysis on large data sets. The ILRI project workflows have been downloaded and used thousands of times and modified for other purposes and re-uploaded.
This new way of doing science, one in which scientists complement hypothesis-driven investigation with a data-driven approach, one in which scientists necessarily let go of control of their outputs to make high-quality data truly and quickly available to the whole global scientific community, gives higher import to the usefulness of scientific findings than either to their originality or to their attribution.
This kind of thinking is poised to go mainstream because the enormous data challenges this new way of doing biological science present are offset by the huge opportunities it offers. Our ability to capture, warehouse and understand massive amounts of data in “creative commons”, where big data become commonly owned data, is changing what science collectively can do and what questions it collectively can ask—and answer.
The disparities I find between (whole) animal scientists and (molecular) animal geneticists at ILRI remind me of the tensions between field and lab geologists explored in John McPhee’s ‘Rising from the Plains’. The biographical focus of that popular book on North American geology is David Love, a legendary rockhound from Wyoming who liked his rocks close up and spent 45 years mapping his state ‘outcrop by outcrop’. Love didn’t have much use for the new generation of geologists who spent their days modelling geologic time and other stuff on their computers, preferring those ‘doing real geology’—out for the day, walking the earth, with a packed lunch and a canteen of water. (I’m a little more generous about the lab side of things, a little less enthusiastic about canteen water . . . .)