This week we have Christa Frodella who will be discussing the technical side of RNA sequencing. RNAseq is a technique used to quantify RNA in a biological sample and has the ability to look at gene fusion, post translation modifications and more. Thank you for joining us, and I hope you keep listening.
This week we have Christa Frodella who will be discussing the technical side of RNA sequencing. RNAseq is a technique used to quantify RNA in a biological sample and has the ability to look at gene fusion, post translation modifications and more. Thank you for joining us, and I hope you keep listening.
Hello, everyone, and welcome to another episode of nerd RX podcast. And today we have Krista fro della, and she is going to introduce us to RNA sequencing. Welcome Krista to the show. Hi. How are you doing?
Christa Frodella:
I'm good. How are you?
Dr. Barkha Yadav-Samudrala:
I'm doing well. Thank you so much for being with us today. And I know you picked the topic RNA sequencing, but before we talk about it, we would love to know more about yourself and your background.
Christa Frodella:
Yeah, sure. Um, so I went to undergrad at Lehigh University. I have a bachelor's in biochemistry. And then I went to Boston University, and I got a master's in medical science. I actually started my PhD at the University of Vermont and bioengineering. I was doing asthma research. And while I was in Vermont, I got really into horses and horseback riding. And I found out that horses were actually a really great model for human asthma. So I talked with my major professor, and he said, if you wanted to go elsewhere to do that, that'd be great. He encouraged me to do so. So that's how I ended up at Mississippi State University now where I'm actually almost done with my PhD. So I start, I started doing my research with equine asthma doing clinical research. And that's actually how I got into bioinformatics and doing RNA sequencing. Unfortunately, during COVID, I had to switch labs. So that's how I got into cannabinoid research and then us crossing paths at the conference.
Dr. Barkha Yadav-Samudrala:
Yes, yes. Wow. Horseback riding. That sounds really cool. Yeah. Well, thank you so much for that introduction. So talking about RNA sequencing, what got you interested in it? And why did you pick that today?
Christa Frodella:
It was completely by accident and how I got into it. Like I said, when I first started my PhD at Mississippi State University, I was doing clinical research with asthmatic horses. And my professor was really interested in why some of our horses were getting more severe asthma than others. So they did lung biopsies, and we were looking at the different the RNA seek of it. And that's how I got into it. And then I also did bioinformatics certificate as also, I also went to a workshop at UC Davis. And I got a lot of training from that.
Dr. Barkha Yadav-Samudrala:
Oh, wow, you have you have been to places. Yeah. That's awesome. So why is RNA sequencing important? And what are the things you would use it for?
Christa Frodella:
So I've used it both, like I said, with the equine asthma aspect, and then also with my cannabinoid research, so we do, can we do research with mice and cannabinoids so low, where you would actually use it is like to look at? Like, how does it change? How does the transcriptome change, like in the healthy state versus the disease state or maybe before and after cannabinoid treatment, and you want to see how it's affecting the transcriptome in which pathways are being activated or inhibited?
Dr. Barkha Yadav-Samudrala:
Okay, and would you mind walking us through the steps that are involved in this entire RNA sequencing process, and just give us an overview?
Christa Frodella:
Sure. So um, if there's, there's actually a few different things you can do. So if you didn't want to do the whole preparation yourself, you can actually send you can get your samples, your tissue or your cells and then just send them out to a company. But what some people do to save cost is they do the RNA isolation, and then they send it out to the company. So basically, from there, they create complementary DNA or cDNA. And then they also require where the company also prepares the sequence Live Library, and then they sequence the whole sample. And then they send you back the raw reads. And then the analysis part is where it gets really time consuming and then comprehend, computationally intensive. And that's kind of where all of my bioinformatics training comes in, really.
Dr. Barkha Yadav-Samudrala:
So what if you don't have the means to send it out to a company? How would you will do this in a lab?
Christa Frodella:
So if you wanted to sequence the entire transcriptome, though, you really only can send it out to a company unless your school has a Um, unless your company or your institution has its own RNA sequencing machine, which I believe Mississippi State does, but we chose to send it out.
Dr. Barkha Yadav-Samudrala:
Okay. Okay, and how long does this entire process usually take?
Christa Frodella:
I think I had results within a week because they just send you they upload the raw reads to, like a Dropbox type thing, and then you can download them on your computer.
Dr. Barkha Yadav-Samudrala:
Okay, and what kind of samples do you send to this company?
Christa Frodella:
You can send tissues or you can send cells.
Dr. Barkha Yadav-Samudrala:
Okay, anything. Okay. Okay, and so for example, in an experiment, you're looking at mice, for example, let's say cannabinoid, how a particular drug is affecting the transcription of certain proteins. So do you like send different samples from different groups after treatment?
Christa Frodella:
Yeah, so you can send like, so you can send like group that wasn't treated versus treated. And then if you have a disease model, you can send no disease versus disease, usually, like anything you do with mice is not a survival surgery, you have to create as many groups as possible. And then when you do your analysis, it's a it's you're always comparing two groups. So like maybe cannabinoid traded versus non combat cannabinoid treated, or maybe a disease group versus not a disease group.
Dr. Barkha Yadav-Samudrala:
Okay, so are there any alternative techniques that you could use instead of RNA sequencing to get similar results,
Christa Frodella:
the closest that you can get is usually like a lot of people do an RNA microarray. So you get this plate, and it has a bunch of genes like probes and then your RNA hybridizes. And then it gives you results, it's definitely nowhere near as good as RNA sequencing now, and it's limited in the amount of jeans.
Dr. Barkha Yadav-Samudrala:
Okay, so when you get the results from the company that provides What are you looking for?
Christa Frodella:
Yeah, so when I do my analysis, I compare two groups, and then we see. So the two things, I'm what I'm most interested into is full change. So like negative or positive compared to each group, and then like a false discovery rate. So usually, you want a false discovery rate of point o five or less, kind of like your P value, your false discovery rate is a little bit more stringent than your P value. And then when I take I take my significant data, you can throw it into a software called ingenuity pathway analysis. So based on the full changes, and the genes and your false discovery rate, it tells you which pathways are more likely activated or inhibited in that certain group. So it gives you a really a lot of information. Because otherwise, if you're if you just have the differences, like all your genes that are significant, and your full changes, you're kind of like going through each one, and be like, Well, what does this mean? So you have a software that tells you kind of like holistically what everything means. But then you can also go through and see if there's any, like genes of interest that pop out for to you.
Dr. Barkha Yadav-Samudrala:
Okay, okay. And in this entire process, and based on your experiment, which step of the wave would you say is the most difficult to learn?
Christa Frodella:
So, definitely the analysis part because I did a lot of, so there's two ways to do the analysis, you can learn how to code. And so you have to learn how to use use like a Linux computers, you have to learn how to use bash, which is like your command prompt. And then you also have to learn like sometimes Python, or C, and then you have to learn our studio to do the statistics. So if you are very coding savvy, you can do all those steps and it's more customizable. But if you are, few are not as strong as coding, you can use something called CLC genomics Workbench by Kaijin. And that's a little bit more user friendly, but it does take a long time. I think one of my RNA data sets took probably two days of constant running.
Dr. Barkha Yadav-Samudrala:
Oh, wow. Oh, yeah. Okay, so bioinformatics, like the analysis part is the most difficult and European? Yes. Okay. And what do you say this technique is user friendly. And I think you mentioned the analysis would definitely require some form of expertise in for you being able to learn right?
Christa Frodella:
Yeah, there it was. It was quite a bit of time of learning before I felt comfortable doing my own analysis, especially if you Do more of like the Linux pipeline workflows, there's a lot of different. There's, there's a lot of different parameters you can change, you can be more stringent, you can be less stringent. And then you have to really be fairly decent at coding because one little mistake in your code and it doesn't work.
Dr. Barkha Yadav-Samudrala:
Does this code change based on your experiment? Or does it remain constant that anyone can run it,
Christa Frodella:
there's definitely accepted values that you can change it like, sometimes you don't produce like, there's not a whole lot of you're not yielding a lot of data. So maybe you want to set more liberal parameters. And sometimes, if you have too much data, you want to set more stringent parameters. So then you have like, a little bit less to work
Dr. Barkha Yadav-Samudrala:
with. Okay, so that is where I think a little bit of experience and coding would come handy. Yeah. Okay, and what are the advantages and disadvantages of this technique in European.
Christa Frodella:
So with RNA sequencing, you can actually use a lot less a little bit less RNA when you send it out. Whereas like microarray, you need a lot more RNA. And then so like I was saying before, you can quantify your gene expression. So that's with microarray, it's not necessarily a quantifiable thing. Something I've also done with my PhD is, you can look at alternative alternatively spliced genes, and that's like something that is gonna probably require coding, I found a program called alternative splicing graph aligner, or ask owl on through NCBI. And I use that to my data as well, you get a lot of data from that. So that's also a lot of parsing through and sitting and figuring out where, where everything goes. Something that I have not done Personally, myself, you can use a deep convolution workflow that's also require encoding. And basically, so if you send out tissue, each cell has like, kind of like an RNA signature. So basically, what the program does is it kind of gives you like the fraction of cells that are in the tissue. So like, let's say you're working with neuro inflammation, it's like, okay, maybe you have like this percentage of immune cells as a percentage of neurons. And that's really helpful in determining disease state or healthy state.
Dr. Barkha Yadav-Samudrala:
Okay, and what about disadvantages?
Christa Frodella:
It's still very costly, I think, the 12 samples that we sent out cost over a little $4,000. And you really, like, if you really want to increase your statistical power, definitely more samples are better. But that's not like if you're working with a limited budget, it's really difficult to do large sample numbers with RNA sequencing. And then it's also computationally intensive.
Dr. Barkha Yadav-Samudrala:
Okay. Okay. And let's talk about the cost involved. I know you just mentioned that it is something really expensive. But is it a technique that any lab can easily set it up? Or? It's very, very expensive?
Christa Frodella:
I guess it really depends on your budget. If you're on a shoestring budget, then probably not. So I actually looked up the cost before speaking with you so. So no, but we went through Nova gene and Nova gene charged us about $100 per sample to create the mRNA library. And then they charged us a little less than$200 to actually sequence each sample. So you're looking at about $300 per sample,
Dr. Barkha Yadav-Samudrala:
or sample Oh, yeah. Okay. Wow. And what about the initial prep? Is that something you can easily do? Or even that is expensive? Oh, no,
Christa Frodella:
it's very easy. So whatever tissue you get, you want to flash frozen it, freeze it, and then put it in RNA later solution, and then make sure when you send it out and has to be kept cold. Okay. So with. Okay, yeah,
Dr. Barkha Yadav-Samudrala:
it's just the company aspect of this is something expensive. Okay. And do you have any fun facts about RNA sequencing?
Christa Frodella:
So actually, if you're working with a very novel organism, you can actually create your own transcriptome de novo. So like with mice, it was what part of the process is you have to blind your your RNA to a reference genome, which you can find on NCBI. But if you don't have it, because you're working with a really novel organism, you can create your own transcriptome.
Dr. Barkha Yadav-Samudrala:
Oh, wow. Okay, and what organisms like what, what animals have you worked with in terms of RNA sequencing?
Christa Frodella:
Um, just horses and mice. Okay. And the other thing another fun fact is let's say you don't Have any data to work with, you can go on NCBI they have something called the gene expression omnibus. And basically, anytime someone publishes RNA data, they have to also upload all their, their raw reads, so you can do your own analysis on it as well. Okay, so with that, because I was just playing around, I've also done analysis on human cells.
Dr. Barkha Yadav-Samudrala:
Okay. That's nice. And my final question to you is, could you suggest any interesting article relating to RNA sequencing?
Christa Frodella:
I was, I would actually suggest, like taking as many classes as you can, because an article is going to be really difficult to understand unless, at least, it was for me, unless you're very computer coding forward, it's just going to be a little bit difficult. So for me, I would I would suggest, like taking as many classes as possible or taking the workshop class, I suggest that at UC Davis,
Dr. Barkha Yadav-Samudrala:
okay. Okay. So does that course is more focused on the analysis aspect,
Christa Frodella:
the whole workflow, so they actually, they get you to learn how a little bit of coding and do the whole workflow, and all different pipelines from beginning to end.
Unknown:
Okay. Wow, that's nice. I'll definitely try to link that down in the description for our listeners. Well, thank you so much, Krista, for being here with us today and sharing your knowledge about RNA sequencing. Thank you for having me. Yeah, it was a pleasure talking with you and listeners. I will catch you next week on another episode of Nerdrx podcast and in meantime, if you have any suggestions or if you would like to come in discuss a technique with me, please email me at barkha@nerdrxpodcast.com and remember its good to be a nerd. Bye.