10x Genomics sells library prep robots for $260k, we took this as an engineering challenge and hacked it together on the OT-2 for $26k.
Whoa thats so cool!
2 Likes
That’s incredible cost savings. I believe that 10X Genomics pretty much uses Hamilton hardware, right? How’s the OT-2 hardware hold up?
2 Likes
It’s not nearly as precise as our microlab starlet. If you watch the pipette cam footage, you’ll see it works surprisingly well. Our sequencing data look great!
3 Likes
I love what you did with the cameras! Kudos!
Recording every pipette step has been on our nice to have list for a long time. Now I feel more inspired to give it a try 
2 Likes
The added bonus of having cameras on systems is being able to troubleshoot from home
2 Likes
Love to hear it! It’s a simple as a Jetson Nano running a gstreamer UDP stream over LAN.
On our nice to have list are tiny computer vision scripts that alert when pipetting errors occur (unequal aspiration in multi-channel, air bubbles when mixing, and pipette homing to precise locations). Great for quickly QCing these 6-hour protocols!
Such a good idea. What cameras have you found work well?
I use the IMX477 sensor with autofocus actuators, this is way overkill. If I build out vision-based control / self-calibration capabilities on the OT-2, I imagine I’ll migrate to a smaller sensor (lightweight data == easier to play with)
1 Like
A Nano? Those aren’t tiny… how did you strap that down?
1 Like
Haha you are a LEGEND! Where did you find a camera ribbon that long?
I should share my Jetson builds
1 Like
here 
signal degradation does happen at this length, it’s all good with a bit of buffering
I wanna to see your nano builds!
This is cool as all get out! Like others, I LOVE the camera integration right there on the pipettor!
1 Like