| Phone Icon 403.210.5275 | Email Icon Contact Us | Resize Text
Post Header Graphic

Dairy Researchers “Making Hay” with Milk Spectral Data

Apart from hurting the caveman’s feelings, reinventing the wheel can be costly and time-consuming. And in scientific research where time and money are in short supply, anything that depletes them may threaten the work itself. Thankfully for members of the Genome Alberta project, Raising Feed Efficiency and Lowering Methane Emissions in Dairy Cows Through Genomics, milk spectral data keeps the wheels of progress turning.

“Milk spectral data (MSD) is a product of the interaction between infrared radiation and molecules in milk,” said Dr. Allison Fleming, Postdoctoral Fellow at the Centre for Genetic Improvement of Livestock at the University of Guelph.

“In regular milk recording programs, all milk comes through milk spectroscopy machines and you get a good data set of peaks and troughs representing the relationship between infrared energy and molecules in the milk.”

Data shows the whey

The data has been used for decades to quantify elements like milk fat, protein, lactose and milk urea nitrogen. 

“The milk tells us a lot about the cow and these various traits. The spectra we obtain for the milk samples are really a window into the animal’s global milk composition and physiological state.”

While farmers use it for management, it creates a large database that researchers can employ for genetic evaluation of all the traits it measures.

“Last year just under 700,000 cows in Canada were enrolled in DHI milk recording programs and were measured every 35 days on average; that generates a lot of data. It’s a great opportunity to benefit from technology that is already in place, working, cheap and has high throughput.”

Because the data contains much information not currently being used by researchers, they are looking at its ability to predict new traits, with feed efficiency and methane emission chief among them.

“These novel traits like feed efficiency and methane emission are very expensive to measure; we can’t measure them on a large population so the milk spectral data lets us generate an indicator to get predictions for larger populations. In doing so, we can increase the accuracy of the genetic evaluations if we can generate more phenotypes.”

Greenbacks and green business

And greater accuracy means better results for producers and more dollars in their pockets.

“Feed costs are a major worry of dairy producers. If we can improve the feed efficiency we will save industry a lot of money and could also reduce land use to help the environment.”

For researchers like Dr. Fleming, it’s heartening to see a lot of interest from the industry as “this is something they have been seeking for a long time. It’s always nice when you can see the real-world application of your findings and know that the work you’re doing actually makes a difference to producers and industry.”

It’s worth noting that they are also making a difference beyond the scope of this specific project.

“We are saving a portion of the spectral data for all cows in Canada so we can go back and apply the prediction equations to a data base. With 3.5 million milk spectra saved, we can apply the developed prediction equations to the database of saved milk spectra and to future incoming milk spectral data to generate predicted phenotypes for these cows.” 

Through it all, the researchers have their eyes squarely on the prize.

“The real point is to find a genetically correlated trait that we can have in a large number of animals cheaply and efficiently,” said Dr. Fleming.  

“The accuracy of these traits won’t be 100 per cent; you can’t predict feed efficiency perfectly from this technology. Ultimately it’s about finding a trait that is well correlated and has a genetic basis that we can use to select for feed efficiency and methane emission in a more accurate manner.”

And with the time and money they save by not reinventing the wheel, they hope to keep their progress rolling along for years to come. 

Dairy Researchers “Making Hay” with Milk Spectral Data

Listen Icon Listen to podcast