Hacker News Reader

Comment by fabian2k | Hacker News Reader

Regarding the question why scientists don't buy the sequencing machines themselves I think a major factor is likely that sequencing is something that everyone expects to be a service and not something you do yourself. I would not even have considered doing this in-house.

At $15 per plasmid there is also no real advantage for typical labs, they're unlikely to save much money here. It could be quicker because you don't have to send the samples, but that likely isn't a big factor here. And having another device that either someone has to run as a service locally or that you have to teach everyone is a rather big effort.

View on HN

I doubt it would be faster in a typical microbiology lab. Every time I've used plasmidsaurus in the recent past I've had results next day. It is a reasonable little bit of labor to prep plasmids for sequencing and then there is a bit of technical expertise required to format the data into a readable format. It's certainly doable, of course! Most lab managers / techs have more pressing things to do than to work on sequencing plasmids. Though I suppose it depends on the scale of your sequencing needs. I found plasmidsaurus to be amazing, and the full plasmid sequencing helped identify some errors that had been propogating in my old lab that never showed up on Sanger sequencing.

"In house" could mean in the institute, though. I never worked anywhere that had in-house sequencing, but better-funded places like Lincoln's Inn Fields did. Those places were big enough to make it cost-effective, and it's hard to beat the turnaround time of a sequencing lab two floors above you.

I work in a large biomedical research institute with in-house sequencing. It's a constant point of contention that many labs continue to send their sequencing out to external companies as they can still be cheaper and more reliable.

Some universities do it in house too

Running the service could involve a more efficient division of labor. Think of Adam Smith's pin factory. I don't know how many manual steps are involved in sequencing. I live in a town with a couple of big diagnostic test labs, and they are very much organized as an assembly line. They hire hundreds of technicians and scientists. There's also modern quality control, which isn't going to happen in an academic lab.

Some labs do but that is a bunch of extra labor that they need to manage. Sometimes some labs do more sequencing than others and they might have more researchers to do the regular sequencing. Many labs are smaller and can't afford to hire so many people or buy so much tertiary equipment. Grad students are already overloaded so they probably don't sequence unless its directly a part of their project

It also depends on what you are doing in the lab. The sequencers for most technologies are not cheap. The prep kits arent cheap. The assays take time. Plus for a lot of labs that sequencer might get used once a year. It just doesn’t make sense to invest to have this tech in house especially when it might end up obsolete in a few years, after you used it a whopping 10 times.

Reply on news.ycombinator.com