The bacteria belowChanges in soil bacteria could help locate mineral deposits
Changes in soil bacteria could help locate mineral deposits
By Christopher Pollon
May 24, 2017
Exploration geochemistry specialist Peter Winterburn is doing pioneering work to discover how genomics could make the task of mineral exploration easier and more economical. Now working with the Mineral Deposit Research Unit at the University of British Columbia (UBC), Winterburn is drawing on decades of geochemistry experience across Africa, Australasia, and South America (including stints with Anglo American and Vale) to explore how deeply buried mineralization can affect changes in the DNA/RNA of bacteria found in surface soil. The ultimate goal: the creation of a handheld tool to genetically sample soil in the field, providing better intelligence in advance of costly drilling programs.
CIM: What got you interested in a mining career?
Winterburn: Like many geologists, for me it started off as a hobby. In South Yorkshire, England, where I’m from, we have Jurassic limestones rich in fossils. By the time I went to university I had the option of either following chemistry or geology. I actually enjoyed chemistry more, but the idea of being out in the open, rather than being stuck in a laboratory really appealed to me, so I did geology instead, though of course never gave up on chemistry.
CIM: What’s the UBC gig like? How is it different from your career spent with big companies?
Winterburn: I’m at UBC as a Research Chair in Exploration Geochemistry. [At UBC] I have the opportunity and budget to apply science to problems I always knew existed in the past, but never had the funding to test. Currently I have a solid budget funded through Bureau Veritas, Acme Labs and NSERC, which allows me to set up research programs to actually go out and test concepts in the field, including some of those far-fetched ideas about how we can find minerals. In addition I can provide training opportunities for future geochemists, an area where demand is greater than supply.
CIM: Can you explain how genomics could be used for mineral exploration?
Winterburn: Bacteria are among the best geochemists in the world, and occupy niche environments. A particular bacteria that is happy living in one environment may well be unhappy somewhere else. They can also rapidly mutate to occupy particular niche environments, so small changes [in the environment] will effectively make one type of bacteria unable to live there, or a mutated version of the same bacteria will occupy that same niche. Sometimes the changes can be as simple as genes in the DNA/RNA being activated within an individual bacteria, which allows it to occupy a particular changing environment. So what we’re doing is looking for the genetic changes which indicate they are surviving in a different environment. The level of change we are looking for is so small it could not be robustly detected using routine geochemical tools.
CIM: Is the study of bacteria to explore for minerals a new area of research?
Winterburn: As far as I’m aware, at UBC, in collaboration with Sean Crowe from the Life Sciences Centre, we are pioneering in this using current available technology. Back in the mid-‘60s, ‘70s and ‘80s, researchers in the U.S., Australia, and Canada sampled exposed ore bodies looking at particular bacteria, and they could see that bacteria populations would change. Historically, much of this work was done in Petri dishes, and you would wait many days to see the results from the sampling, which came from exposed ore bodies. What we’re looking at now is exploring environments where ore bodies are buried beneath transported material. With modern sequencing techniques, you don’t have to grow anything in a lab anymore, you can take soil and sequence the genes, identifying those changes that are occurring above the mineralization. Finally you need to verify that those changes are only occurring above mineralization. One of the downsides of genomics is that it produces huge data sets, so you must have the computing and statistical power to analyze the data. In the same way that the price of doing the genomic analysis itself has plummeted over the last decade, the ability to work with these massive data sets and statistically manipulate them has improved at a phenomenal rate. The end point however would be a simple handheld tool looking for very specific genes.
CIM: The sampling of bacteria is all happening at the surface, right? You’re not looking at drilling and analyzing the bacteria in cores as well?
Winterburn: We are looking primarily at surface sampling. Drilling is expensive, it requires permitting, you have to build access roads, and the [drilling] is environmentally invasive. But with surface sampling, you walk across the ground, digging a shallow hole every 50 metres or so to take a sample.
CIM: What do you need to demonstrate through research before these genomic tools can be reliably used in exploration?
Winterburn: We are currently doing a proof of concept exercise. We have samples sitting in freezers at -80 C collected as part of my ongoing research programs. We will run those samples through the sequencing system to understand the types of responses we are getting. We’ll take that proof of concept to industry, with the potential to get more funding, and then take it forward for proper research on a greater variety of deposits. That will allow us to establish whether [the results] are just specific to the deposits we tested, or whether it’s more generally applicable, in terms of an exploration tool. That way we will avoid one of the downsides of much geochemical research done in the past on buried deposits. Often an orientation survey at one deposit produces interesting results, a big song and dance is made about it, and it subsequently turns out that the particular technique only worked at that one deposit.
CIM: The samples waiting to be analyzed, they have been taken from all over the world?
Winterburn: We have soil samples from a volcanogenic massive sulfide (VMS) ore deposit on Vancouver Island, Highland Valley and Woodjam copper porphyries, central British Columbia, a kimberlite from the Northwest Territories (which is interesting, as there is permafrost and no trees to assist with element mobility), as well as a very arid, super-saline area in the Atacama desert of northern Chile. All of these areas are concealed by young transported cover.
CIM: So in the future, a mining exploration company would be able to genomically analyze surface samples in the field, in areas they already think are promising?
Winterburn: Yes. The concept we are chasing currently is target evaluation. Once you’ve selected an area of interest, perhaps from geophysics or geological observations of outcrops, you would collect a suite of soil samples across the surface. The difference is you wouldn’t send those samples to a lab. Our idea is that you can identify very specific changes or an activated gene in the DNA/RNA, and build a detector that is very specific for those particular genes. It would effectively be a palm-sized device, you would take a soil sample and do a DNA extract. You’ve seen those little diabetes test kits? It would look like that. You would take some of the solution from the DNA/RNA extract, put it on the device, and it would tell you if those particular genes are present in the sample or not.
CIM: And whether they are present or not, this could provide the information needed to advance a more in-depth examination of the site, like with drilling?
Winterburn: Yes. The way I like to think of any phase of exploration is, each piece of information you collect adds another layer of evidence to help you make an informed decision. The biggest mistake that companies can make in exploration is to focus on one piece of info and drill it, with variable success. The more layers of evidence you can gather, the greater your level of confidence in the decision to either drill or walk away.
CIM: At what point do you think someone will have a handheld device to do this kind of field testing?
Winterburn: To be optimistic, someone could be out in the field doing the extraction and analysis in three to five years, obviously depending on the results of the research.
Combining geostats, AI and VR to find gold deposits
Jean-Philippe Paiement explains how SGS has combined historical geological data with machine learning and virtual reality tools to identify and visualize drilling targets