Ask Me Anything: Fern Pearston


Userlevel 4
Badge +2

Join Fern Pearston, beef genetics data analyst at Genus ABS

for our next Ask Me Anything (AMA) session

 

When: Thursday 9 June 2022 at 12:00 BST

Where: Questions and answers posted in this discussion

Introducing our guest: Fern Pearston

 

Fern started her career as a Data Technician at Scotland's Rural College (SRUC) and from there moved to the Agriculture and Horticulture Development Board) AHDB to the role of animal genetics manager in 2014. 

 

During her time at AHDB she helped roll out genetic evaluations for new traits in the UK, further developed AHDB Dairy’s online tool, the Herd Genetic Report, and assisted the UK herdbooks to maximise the use of genetic and genomic information. 

In November 2021 Fern started at Genus as their EMEA Beef Genetics Data Analyst, assisting the T14 and T15 genetic evaluations for the use of beef on dairy and also supports the development of new programmes for the business.

 

Submit your questions

 

Got questions about data, working in a similar field, moving from data technician to analyst? Whatever your question, submit your questions below and join us on Thursday to be a part of the discussion.

liammccaffrey 1 month ago

That’s all folks. Our hour is up. Huge thanks to Fern for joining us today and giving us such insightful answers and thanks to you for all your questions.

Next week, an AMA with one of Peak’s graduate data scientist, Helen Craven. Look out for the event announcement on here.

View original

15 replies

Userlevel 6
Badge +2

What’s the most interesting thing you have learned about the agriculture industry from your work?

Userlevel 3
Badge +1

What programming languages do you use in your job and what for?

Userlevel 3
Badge

What’s the most interesting problem you have worked on?

Userlevel 3
Badge

What made you interested in working in the agriculture industry?

Userlevel 4
Badge +1

How did you get into your field?

Hi Fern! 👋

Moving from academia into an industry role really highlighted to me how essentially the same topic is sometimes approached using very different methods.

What differences do you see in working in genetics in academia vs industry?

Userlevel 4
Badge +2

Big welcome to Fern. AMA has started now and she’ll be with us until 13:00 BST 🙂

Userlevel 1
Badge

What programming languages do you use in your job and what for?

Currently I use R for almost all of my data work.  I use this to generate descriptive statistics of our data to understand if there is incorrect information which shouldn’t make it into our live database and also to make snazzy graphs for visualising the data for my colleagues to diseminate the information further.  I am using Python scripts for data edits too but am less well versed in this language (following a script a colleague has created for one of our projects).  In my previous role I used SQL Server daily to interrogate data and set up procedures for processing data 

Userlevel 1
Badge

What’s the most interesting thing you have learned about the agriculture industry from your work?

I didn’t come from an agriculture background myself so have made my way in through my jobs.  So I think the most interesting thing I’ve learned from working with farmers over the last 8years is all of the hats they need to wear on a daily basis (understanding animal health and welfare, breeding, financials, health and safety as well as many more!) and how much effort and passion goes into providing food for the country.  It was really incredible to go from a consumer to being someone who helps farmers provide this service to the public and humbling to see how much effort goes into making sure we have milk on our supermarket shelves.  

Userlevel 1
Badge

What’s the most interesting problem you have worked on?

In my previous role working for the dairy levy board, we provided a Herd Genetic Report for all farmers who diary herd data flowed into the national genetic and genomic evaluations.  Through this report, farmers could see their milk cows’ genetic potential and make decisions such as which to breed replacement cows from (those with the best genetics which they would want continuing into the herd) and those which don’t have the right genetics after all and so might be put to a beef bull so their progeny won’t enter the herd and instead would flow into the beef supply chain instead.  The gap in the tool was that the famer had no idea of the genetic potential of their youngstock (the calves or young cows yet to calve) so when making breeding decisions they were making them blind.  I delivered an update which looked at all the youngstock in our data and calculated their genetic potential.  It doesn’t sound like much but it was a huge data set and the method for calculation depended on what parent information we had for the calf and whether it was a pure breed or a cross between breeds.  This is now being used by UK dairy farmers to better inform their breed decisions, after all, if their breeding policy is successful their youngstock will be some of the best genetics in their herd so understanding this vital for the continued genetic improvement in their herd.

 

I moved jobs in November last year so am just getting stuck into the projects in my new role but we are looking at dairy cross beef animals in the beef supply chain.  These animals have started as a bi-product of the dairy herd but through my current work we are looking at genetics to improve the efficiency and carcass quality, turning a bi-product into a valued product, wanted within the supply chain.  It links very nicely to my last role and is an exciting place to be with the committments being made towards net zero which these traits will play a role in throughout the supply chain.

Userlevel 4
Badge +2

Two questions from me:

  1. Agriculture (from an outside perspective) doesn’t intuitively seem like the most obvious adopters of data science/analysis. Can you tell us about what motivated Genus ABS to get started with leveraging data and how they structurally (e.g. software, organisational structure, data literacy of decision makers etc) supported it?
  2. We hear a lot about dirty data, data silos etc. How do you think overall the agriculture industry is doing with managing their data so they’re usable for data practitioners like you?  
Userlevel 1
Badge

 

Hi Fern! 👋

Moving from academia into an industry role really highlighted to me how essentially the same topic is sometimes approached using very different methods.

What differences do you see in working in genetics in academia vs industry?

Hi!

I think the big difference is seeing the results of my work being applied and used in the real world - making differences to people’s livelihoods and businesses. Being a data technician, I often did the leg work of the analysis but never presented the work in meetings so it was nice to take ownership of my work and see it implemented.

By presenting my work more in an industry environment also encouraged me to break down my understanding of topics - presenting to audiences and colleagues who don’t have the same background as me made me rethink how I explained the details in a way that made sure everyone understood the information and was brought along on the story I was telling.

Userlevel 1
Badge

What made you interested in working in the agriculture industry?

How did you get into your field?

I’ll answer these two together as I think they linked for me.

I was someone who didn’t quite know what I wanted to do as a job when I finished school but knew I loved maths so went off and did a maths degree.  At the end of that I didn’t want to be a maths teacher, go into finance and wasn’t smart enough to be a spy at GCHQ (no living out Spooks in real life for me).  However, during my final year of my maths degree I studied areas of maths applied in biology and this sparked my interest.  It led me to apply for a masters in Quantitative Genetics and Genome Analysis and although I hadn’t studied biology since Standard Grade (GCSE for the non-Scots) I was accepted.  This allowed me to explore the area of animal breeding and genetics which really appealed to me.  My master’s thesis was looking at including maiden heifer fertility in the current Fertility Index based in the SRUC office in Edinburgh.  I knew after my masters that I wanted to get a foot on the career ladder, working on projects within this area.  I was fortunate enough to be offered a role in that office as a data technician and from there, it was being in the right place, at the right tim, having the right connections and taking opportunities when they are offered that’s led me to where I am today.  I wasn’t the best geneticist but understanding the data and using it to provide insights has became more my role and driver.

Userlevel 4
Badge +2

That’s all folks. Our hour is up. Huge thanks to Fern for joining us today and giving us such insightful answers and thanks to you for all your questions.

Next week, an AMA with one of Peak’s graduate data scientist, Helen Craven. Look out for the event announcement on here.

Userlevel 1
Badge

Two questions from me:

  1. Agriculture (from an outside perspective) doesn’t intuitively seem like the most obvious adopters of data science/analysis. Can you tell us about what motivated Genus ABS to get started with leveraging data and how they structurally (e.g. software, organisational structure, data literacy of decision makers etc) supported it?
  2. We hear a lot about dirty data, data silos etc. How do you think overall the agriculture industry is doing with managing their data so they’re usable for data practitioners like you?  

You’d be surprised just how much data is out there in the agriculture world!  Genus ABS is a bovine genetics company so make money through sourcing high genetic bulls to sell straws of semen from to dairy and beef farmers. However the role of genetics companies are changing and instead of just selling straws of semen, because the dairy industry specifically is working towards improved fertility which means less straws are required per pregnancy profit margins are decreasing.  By leveraging data and gathering our own real world data we are able to provide additional genetic progress to our customers and not rely on straws as heavily to deliver a profit for the company.  Services are becoming more common now as way of delivering for our customers.  To provide these services we have dedicated teams working on technical services (which would be provided on farm) and also in-house services, such as genetic tools to offer our customers for further insights and understanding of their herds as well as proprietary genetics indexes.  With having specific teams dedicated to this, it is well understood by the decision makers that the data we are investigating in is vital to current and future services we would like to deliver. 

I hope this has answered most of your first question but let me know if you would like me to expand on anything further.

 

I think the difficulty is that there are a lot of providers and technologies which collect data in the industry and this data isn’t always compatible, animals are identified differently says, or technologies just don’t talk to each other.  Its frustrating for farmers that use the technology and can make life difficult for data practitioners.  However, I am aware of data standards set by ICAR (International Committee for Animal Recording) which is some standardising data recording aspects and making it easier to combine this data from multiple sources. 

Regarding dirty data, I think the more that we can provide services to the industry that shows the value that can be added to the business by having access to good data, the better the data recording will be on farm but we need to show that value - there’s a lot to do when running a farm business and sometimes data quality isn’t the top priority but for those who do know that power of data, their records are really good.

Reply