Programming The Beckman Coulter Biomek Fx Robot
2
3
Entering edit mode
11.6 years ago
Weronika ▴ 300

Our lab has a Biomek FX "Laboratory Automation Workstation" and I'm going to need to program it.

What we need done is basically moving/combining samples from arbitrary wells in several 96-well plates to arbitrary wells in a 24-well plate, for several plates. (The point of all this is combinatorial pooling with error-correcting binary codes for deepseq samples, not that it matters). I already have a python script that defines all the steps that need to happen (i.e. which source plate well needs to go into which destination plate wells), I just need to figure out the correct format for a Biomek file with the same information. I know the Biomek has an option of loading a file that will tell it what to do, but the manual is pretty long-winded and confusing to read from the point of view of someone who doesn't need to know how to use machine, just to make the file.

Has anyone here done this? Is there an example file you could give me?

• 8.1k views
2
Entering edit mode

I can't see the qualitative difference between this question and a question about converting Illumina NGS file formats between different encodings or writing code to pull down a particular chunk of the UCSC database. A lot of questions posted on this site and considered within bounds are not research topics; they're extremely specific requests for help with a practical engineering problem that arose from a bioinformatics analysis. I still think there's a long way between those tasks and installing printer drivers. (And my thanks to those whose main job is installing printer drivers).

1
Entering edit mode

This falls outside of what I consider to be a bioinformatics research problem. Closing for that reason.

0
Entering edit mode

@neilfws: Oops, sorry. I figured biology + programming = bioinformatics.

0
Entering edit mode

I think this one is similar to a question posted by Pierre. This question also got 2 votes in approx 8 hours, that indicates there is some level of interest. It will be nice if we could reopen this.

0
Entering edit mode

If someone else wants to reopen, go ahead. Personally, I think that when we define bioinformatics as "anything with a plug" (as one of my colleagues says), it devalues the subject and strengthens the perception that what we provide is little more than IT support.

0
Entering edit mode

As this is a very specific commercial platform, I doubt you will get a useful answer here. You'll have more luck with (a) reading the manual properly or (b) sending a support request to your supplier.

0
Entering edit mode

Perhaps adding a definition of what is considered bioinformatics to the FAQ would be useful to prevent this kind of confusion, then. Still, I think a question about programming, not just using or fixing, a complicated biology-specific machine is pretty far from "anything with a plug". And yes, I'm aware my chances of getting an answer aren't high, but I thought it was worth asking. If I don't get anything, I'll post the information myself once I figure it out, and maybe someone else will find it useful someday.

0
Entering edit mode

We have several Fx stations at our site... but why on earth do you directly want to python program it? The accompanied Beckmann programming FX software has so far fullfilled all our needs. Maybe my ignorance...

0
Entering edit mode

We're doing "combinatorial pooling" - basically we need to program it to combine samples into larger pools in a complicated and non-regular way. Luckily it can take simple plaintext input that defines all the operation, so I can generate that file in python, I just needed to figure out the format (unsurprisingly it didn't turn out very complicated).

2
Entering edit mode
11.4 years ago
Weronika ▴ 300

It turned out very simple, which really shouldn't surprise me. Comma-separated fields: source plate, source well, destination plate, destination well, volume - like this:

Source_1,A1,Destination_2,A2,100
Source_1,A1,Destination_2,B2,100
Source_1,A1,Destination_2,B3,100
Source_1,A1,Destination_3,A1,100
Source_1,A1,Destination_3,A2,100
Source_1,A1,Destination_3,B2,100
Source_1,A2,Destination_2,A1,100
Source_1,A2,Destination_2,B1,100
Source_1,A2,Destination_2,B2,100
Source_1,A2,Destination_2,B3,100
Source_1,A2,Destination_3,A1,100
Source_1,A2,Destination_3,B1,100


Now if only I could get the Biomek software not to hang when given a 600-line file...

1
Entering edit mode
11.4 years ago
Berwyn ▴ 10

Hi Weronika. I look after the Biomek liquid handlers over in the UK and came here via Twitter. Just saw your post and am glad you worked the file out. However, the Biomek software 'transfer from file' step should not crash the software with 600 lines. Thats only the equivalent of 6 plates and we have many methods that use a lot more than that. This suggests there is somehting else wrong. You may want to get in touch with Beckman US for some assistance. Let me know if there is anything I can do to help.

0
Entering edit mode

Yes, the tech in my lab who's doing the actual robot handling should be talking to Beckman about that problem sometime this month. Good to know you think it should be possible to solve! Thanks for the note.