Kristin Ellis recently visited the headquarters of Biomarin Pharmaceuticals in San Rafael, CA. There, she had the opportunity to speak with Elaine Phan, a Senior Research Associate in protein biochemistry at Biomarin, who has been using the OT-One PRO to run protein quantification assays. (Conversation has been edited for clarity.)
Biomarin’s Elaine Phan using her OT-One PRO robot.
Kristin: I’m very excited to hear how Biomarin has been able to utilize the OT-One PRO, but first, can you tell me about your overall experience with Opentrons so far?
Elaine: It’s been amazing. You started sending me protocols and I’ve been able to just do them! What I’ve done so far is just taken notes from what you wrote me regarding automating a Bradford Assay, then looked up the details of how to write the protocol from your website – I think there’s a webpage where anyone can learn it?
Kristin: Do you mean the open-source API documentation on docs.opentrons.com?
Elaine: Yeah, and then there’s also an official Python learning school Artyom suggested: SoloLearn. In the beginning, though, I learned your JSON protocol writing from what Will wrote on the blog, and that made it so much easier to learn the Python stuff.
Kristin: Oh really? Why do you think that was?
Elaine: Yeah, because when I was looking at the Python it seemed like…a lot harder. But from the JSON instructions, I was able to go through the documents that you guys wrote and figure out how to add and write the different commands. That was really exciting to learn! Once the hardware was here, I was able to put the machine together, and go through the code that was already available for the Bradford Assay and do some optimization – spend some time really fine tuning it. For me, because I’ve been doing this for over 10 years – pipetting, I mean – this is gonna save my arm a lot of wear and tear.
Kristin: Seriously! So can you tell me a little bit about the work your group does?
Elaine: We focus on protein purification – we’re a protein science group. We have a huge pipeline, and I’m at the early research stage. We get lots of fractions, and we need to determine purity on all the different fractions. This means we need to run gels, do Bradfords, and then pool everything that we need. Next we make enough material that can be put into pre-clinical samples to show efficacy of the drug. Basically, we have to make enough material on this end to give to the other groups to do their experiments. Once those are done, we put the drug into the mouse models and see if they work. Having an Opentrons robot makes that cycle a lot quicker and a lot more useful.
Kristin: So where are the proteins that you’re testing coming from initially?
Elaine: We have another group that does cell culture. They grow the cells for me. We have mammalian cells, and also E.coli. Right now, I’m working on E. coli, and it’s expressed with a His-tag on it, which is pretty straightforward. You have a His binding column, it binds, you elute it, those go into the fractions, and then you can see where your protein is located.
Kristin: So you can see the expression of the protein, essentially.
Elaine: Yeah, so… it goes like this. First, I have my loading step. I’ll load my sample in, and whatever doesn’t bind to the column will come out, so all the junk that I don’t want kind of flows through. Then you have a wash step which rinses the columns, and the material elutes through an imidazole gradient, and eventually only what I want binds to the column. I would collect what does bind in a bunch of fractions, and then I analyze how pure it is by running a gel. I need to run a Bradford to quantify all the proteins to show recovery. Then I would run a gel on each fraction, or pool a few fractions together and run a gel, and then see where the protein is by Western. So that’s what we do on a regular basis!
We eventually want the OT-One to do ELISAs. We have to show activity, etc. I don’t personally work on the ELISAs, but other people will, so once they have the workflow finalized I can get the ELISAs and other activity assays from them and try to put them on the robot.
Kristin: Yeah that’s one thing that we’re actually working on right now with a few other groups – we’d love to automate all different kinds of ELISAs. Not just single-plex, but MSD plates, Luminex, as many things as we can develop. So that’s a thing that we certainly want to move into.
Elaine: So when I decide to start doing that I can ask you more questions!
Kristin: Definitely! That’s that’s what we’re here for, and why we’re building this community!
So it’s been like a month and a half that you’ve been using the OT-One PRO. You said you’ve been doing biology and manual pipetting for 10 years. Then you got the OT-One PRO, and you’ve started learning how to code and how to build experiments around automation – how has that changed your experience of working in a biology lab? You know, just working on coding your experiments versus following along by hand?
Elaine: For me it’s opened my eyes totally. There’s this whole world that I didn’t even know was possible for me to actually get into and understand, and it’s just made me really excited. I hear you guys talk about this machine like a personalized computer for experiments, and for me it’s also almost like my assistant. Since it can run a lot of my Bradfords and other assays that I do on a regular basis, I have so much free time to be reading papers and understanding the science behind a lot of the work that I’m doing, and ways to improve our experiments. Not only that, it’s like once you realize there’s this way you can code, it just opens doors to so many ideas I can’t even explain right now. I can’t begin to describe how many ideas go through my mind that I just am not able to express yet because there’s still so much to learn.
And the fact that there’s so many possibilities of what you can do with this machine – you know like, using different pipettes. Like even though you guys said the robot may not work with a Rainin, I still tried it, you know? I was like, they’re not going to stop me, I’m just going to try this!
Kristin: We’re not going to stop you for sure! That’s one of the reasons for having the Opentrons platform stay open-source so that when we say “Well, you know, we don’t do that on our end” you can figure it out!
So I know you think the code part is cool, and you’ve already started adapting the hardware – what do you think of the hardware modularity and the fact that you can just kind of mix-and-match stuff?
Elaine: Well that’s been really awesome because I can actually put anything on the deck! I didn’t even think of these well things [points to tip racks with Beckman Coulter reservoirs] before because we would have disposable plastic ones, but now I can think of like a million types of fluids I could put in there.
Biomarin repurposed their Opentrons tip racks into Beckman Coulter reservoir holders.
Elaine: I also feel like I want to meet other people working on this to understand. I mean, I know you have small version of it on your Slack channel, but I think if I actually meet people and explain what I’m doing and they explain to me what they’re doing, these ideas and will really come together.
Kristin: So it sounds like you’re really interested in the community aspect of it too. Do you think, as a person having so much experience in science, that this kind of community is very important to develop and build around?
Elaine: I think so, especially for me, because right now I’m the only one using it in my company. So I really don’t know who I would talk to besides you guys to understand the broader aspects of what the machine can do. For instance, just today you mentioned that you have people working with magnetic beads and the Amazon Alexa and all of that. And actually, I mean, I’m just at the beginning level coding, right. So how much more coding can I do? And how do I learn that without feeling uncomfortable asking?
Kristin: That’s kind of the idea behind our Protocol Library. We’ve recently relaunched it, and part of the reason for us saying, for instance, “We partnered with Biomarin on the Bradford Assay” is so that people can see that protocol in the library, and then if they have a question about it they can hop onto the community channel and say to you, “Hey, how did you develop this?” Also, the more users we have, the more protocols we can add. Then, you can download them to look at the code and start to understand it, the same way you did when you first started.
Elaine: Yeah, that will be so helpful. Being able to go to them and ask questions after you’ve seen it makes it so the Protocol Library also provides kind of a starting point for the conversation. We’re actually planning to bring in an Andrew and I’m starting to realize that you guys can do all of the things they can do, maybe, if we can just download the codes. They have these ideas that are great, and theirs is just drag and drop which is really easy for people who don’t know how to code…but if I can learn the coding I can just get the OT-One to do it, and it’s way inexpensive, which I’m really excited about.
Kristin: That perspective is actually really interesting! So what if you look back towards the beginning of your career, and let’s say you were to start out with a tool like the OT-One in your arsenal – say like, this and a 3D printer – what would have been the possibilities?
Elaine: That would have been amazing! I would be able to write so many of my experiments with this, now that I think about it. So many things, especially the fractionation…and you know, there’s days when you just have to stay really late to get something done because it can’t wait, and it can’t go in the fridge, or your proteins will precipitate so you just have to get it done to a certain stage so you can stop it or freeze it. In so many of those circumstances, this would have been awesome to have as an assistant.
Elaine’s “assistant” hard at work.
Kristin: How did you decide to start with the Bradford Assay? Has that just been a big need in your lab?
Elaine: Yeah, it was something that everybody runs, and it’s just that basic thing that everyone coming in will learn, because everybody needs to know what the concentration of proteins are all the time. So that’s why we were like, if the machine can run this Bradford, then it can run ELISAs, it can do gel preps, it can do a standard curve. The Bradford is just a basic pipetting assay that everyone has to do when they start in the lab. If you can do those really accurately, then you can probably do all the assays that we need done in the lab.