Planning Genomic Projects Are Like Furnishing An Apartment

note: not an endorsement for IKEA, commercialism, disposable lifestyles, or the idea of 'needed' furniture. IKEA is also not the most inclusive idea to make an analogy out of, but it suited my needs more than any other that I could think of. If you can think of an alternate, I'd love to hear it in the comments section!

I’ve been trying to come up with a way to describe my current difficulties with my project, and I came up with a pretty good idea while talking to my boyfriend about the issue. In the end, he didn’t need this analogy because he studied science and was familiar with the basics, but trying to explain my project using an analogy was a really useful practice. And then I became obsessed with it. The best way to explain it is to define a few variables and dive straight into the madness. So forgive this foray into analogy, I just hope that it works!

I also hope that this can help my friends and family. Before I begin, I'll define a few terms quickly:

• A genome is all of the DNA sequence in the nuclei of your cells. If your DNA is the blueprint of you, the genome includes the plans and the paper that it's printed on. Not all of your DNA is coded for proteins, some of it is cutout and some is just there for structure.
• Sequencing is how we figure out the patterns of what makes up DNA, and it keeps advancing. The old method, popular in the '90's, was called Sanger sequencing, and is quite expensive, but fairly accurate. methods that came after Sanger sequencing were called Next Generation Sequencing (I use Second Generation Sequencing, or SGS), and there are new advancements now that are sometimes called Third Generation Sequencing. Illumina and PacBio are SGS technologies and vary a bit in the way it reads DNA, such as long or short. Costs vary.
• Once you get the sequences for an organisms whole genome, it's a lot of data, like if you had a library card number that was a million numbers long. Except that you don't get the genome sequence in one long read, you get it in pieces, either small or large or both, depending on the method you pick. You have some overlaps, so you have to align these sequences into big enough pieces (scaffolds) that you can eventually make a reference genome. That's called the alignment, and after you align it, you try to figure out what parts of the sequence do what, which is annotation. Once annotated, a genome is considered complete until it is updated.

Note: If you're unfamiliar with my project, go to my previous post or the BINGO website for details, otherwise this may not make sense!

Planning Genomic Projects Are Like Furnishing An Apartment

Goal:  Construct a genome of an organism as to suit your needs and on budget, just like furnishing an apartment as necessary and on budget

IKEA is Second Generation Sequencing, and there isn’t really an option here, because Sanger sequencing is hella expensive (think really expensive reproduction furniture on a shoestring budget), and third generation sequencing (3D printed furniture) isn’t nearly error-proof or cheap enough yet for non-model organisms.

Model of the furniture is the sequencing company that I pick, and the company in a way determines the method of sequencing as well, just like how if you pick one style of sofa, it may only be available in a few colours as compared to a different.

Size, colour, and style of the furniture is the method of sequencing, and that the prices vary based on trends, taste, and materials. The amount of coverage of a sequence increases with increasing quality of the reads, just like a larger piece of furniture costs more than its smaller counterpart. But if you only need a two person sofa, you probably don’t need to blow your whole budget on anything bigger. It’s the same for sequencing. If 40x coverage of long and short reads is enough (a 2-seater sofa with a stable frame, arm rests, and a nice fabric), you don’t need to splurge on 100x coverage of long and short Illumina reads with a PacBio long read (a 5 seater-sectional in leatherette). It’s about balancing quality with budget and intended use.  

Building the furniture is the assembly & annotation of the genome. Some sequencing companies offer assembly (like furniture that is prefabricated and just needs to be unwrapped), so all you have to do is annotation, where you use other methods to see what part of the genome is being transcribed or expressed. This is additional assembly is costly, and may not be exactly what you want, so you can assemble your scaffolds into a genome and then annotate it for yourself (the bare bones IKEA furniture that need nuts and bolts and hex keys and maybe even a drill).

As for the ease of assembly, the more I want done for me, the more I pay. From the furniture angle, I could pay a handyman or call a friend to help me put it together, or I can do it alone. I can use the instructions, or I can prefer to build it from scratch without unwrapping the instructions. This is the same for genomic projects. I could work alongside a bioinformatician or take a lot of course, or I could use an expensive but intuitive computer program to help with assembly. As for from scratch, the genome assembly version is using a program that requires coding knowledge, and even making your own assembly algorithm and/or program from scratch.

--------------------------------------------------------------------------------

Once you’ve built the furniture, that is, once you’ve assembled and annotated your genome, there’s still more to do – you’ve got to use it and make it a functional part of your projects via analysis and improvement, just like when you begin to add personal touches and more furniture or decor to a room.

I hope this has made sense, because I’ll go further into my own situation now, and I hope that it makes sense!

I’ve gotten hold of a nice three room apartment with a four-year lease, the neighborhood picked out for me and ready to move into. Now, unlike a lot of Dutch apartments, it does have a floor - some nice laminate that was installed properly - and a sturdy ceiling complete with lighting. This apartment, empty but full of promise, is my PhD project. The neighborhood in question is the BINGO project, and the floor and ceiling are my graduate school and the department that I’m working in. It’s a perfect place to live for the time being, now I just have to make it livable.

Not my apartment - from Adam Ekberg's exhibition, Next to Nothing

Time to buy some furniture!

This apartment we’re talking about here, it’s nice, with three distinct rooms. Three rooms, to be precise (ignore the bathroom, it’s the only fly in the ointment of this analogy). Three rooms that are distinct, and need furniture that will suit my needs, tastes, and budget. If you are on a genome project and you have one organism that you’re studying, or even one strain, you’re in a studio apartment. Your furniture can have multiple functions – a bed that is a sofa, a dining table that is also a desk, etc. However, my floorplan is a bit less flexible, and I need to furnish accordingly. Because I have three species that I’ll be assembling genomes for. As a reminder, I’m going to be studying a parasitoid wasp (Trichogramma brassicae), a predatory mite (Amblyseius swirskii), and a predatory bug (Nesidiocoris tenuis). I will not be comparing them, but rather doing population-level genomics - how do different populations of these species differ, especially to the commercial populations that are sold as biological controls.

So these three species, these three rooms, are going to need separate sequencing methods and assemblies, and I have to decide what method I’m going to go with. When it comes to genetic sequencing and genomic assembly, there’s a plethora of choice out there, just like IKEA furniture, but you need to choose wisely. Added to this thought exercise is that I will not be going to IKEA myself (akin to using the machines myself), but rather submitting online orders and awaiting delivery, hoping that everything went well beforehand (send my prepared samples and wait for sequence data to be send to me, with quite a bit of time between).

From the LERHAMN series, with Nesidiocoris tenuis inset

First room: The Kitchen | Nesidiocoris tenuis

This is the most straightforward of my choices to make, and probably won’t go in to future analysis. I’m going to buy a table with some chairs with this room and call it a day. I’m not limiting myself too much, but I’m also not going to lose sleep over what else I can do to furnish the kitchen. Now, the species itself isn’t that well studied (genetically), and there may be some interesting insights down the road, but for now I’m banking on not needing more than a fairly solid reference (or near-reference) genome with annotation and some comparisons later on.

From the ÅRVIKSAND series, with Trichogramma brassicae inset

 Second room: The Bedroom | Trichogramma brassicae

It’s kind of straightforward furnishing a bedroom, the main piece of furniture required is right there in the name. Bedrooms need a bed! And I prefer my beds to be robust. But do I want to go overboard in furnishing the bedroom? In the case of my actual project, with Trichogramma brassicae, not necessarily. It’s a fairly well studied species and is used all over the world, and the genome I assemble will be used for marker selection on another project in BINGO that will monitor populations post-release. That’s important for making sure that biological control species do not have a negative effect on the environment, but as for further analysis, I think there’s a more interesting story to tell...

From the KLIPPAN series, with Amblyseius swirskii inset

Third room: Living Room | Amblyseius swirskii

Furnishing a living room is crucial to optimum relaxation. It’s where you relax after work, where you have friends over for a party, and where you Netflix and chill. Very important to furnish. In the case of my study species, this little predatory mite is one of the most lucrative biological control agents out there, yet it’s genetic history is murky at best. I don’t want to give away too much before I being to crack open the genomic analysis, but let’s just say that the term “Achilles heel” will likely come up in my research proposal. I mean, what would you expect to happen if you isolate a small population for hundreds of generations without any immigration or exposure to varied conditions. My living room is this little mite, and I believe that over time I will add to the sofa with additional pieces, and it will eventually become the big story that I tell in my dissertation.

A more detailed living room in the future? Time will tell

But my fictional apartment, it’s quite empty at the moment. I’m sleeping on the figurative floor. And that’s because I haven’t picked which models of furniture I want (which company I decide to go with), what size or colour of furniture I want (which method of sequencing), and how much I want to build alone and how much I want built for me (method of genome assembly and annotation). I also need to keep in mind that the genomes I generate will be used by different projects with their own goals and time lines, so it's a balance of what I want to do and what others need.