Category Archives: Arduino

The Budget Biotech

The bench beaver is always watching you

I have been toying with the idea of starting my own biotech startup in my garage, and want to determine the lowest amount of upfront investment it would take to set up a functional molecular biology lab.

OK, a little background to help the discussion. Straight out of grad school I took a job with a startup biotech developing strains for fermentation of high value small molecules. When I arrived the lab had a few tables and some basic equipment, but being that an electrical engineer did the initial purchasing it wasn’t quite ready for much biochemistry. So my first job was to outfit the lab with the necessary equipment and consumables to get things moving. I quickly realized how expensive laboratory equipment and research reagents were without the generous academic discounts I had grown accustomed to. We pieced together a mixture of new and used equipment, took advantage of lab startup promotional deals, and ultimately assembled a fully functional molecular biology lab on a pretty modest budget.

Taking what I learned from setting up my first biotech lab, I know I can do better. How much better? I think it should be possible to assemble a functional molecular biology lab for under $1,000. But wait you say- that is just enough to buy a decent set of pipetters- how could you set up an entire lab for that?

Precisely. That is the heart of the challenge. To create a budget-minded biotech that is functional, but doesn’t cost an arm and a leg. As I wrote up in a previous article, “biotech shouldn’t be so dang expensive!”

I think it all depends on how you define the core functionality. To me, the minimum functionality needed is to clone genes and express proteins. I am focusing on the steps from source DNA to expressed recombinant protein. So basically from PCR through cloning in E.coli, to expression of proteins.

Things I am not including in the $1,000: Analytical (in my case HPLC or LCMS), basic infrastructure like tables, stools, computer, etc., and little things like surge protectors or extension cords. And I may take the liberty of culturing my own competent cells to cut down on that cost. Vector backbone will be an open-source design to get around IP conflicts and keep things cheap. We’ll see how far we can stretch $1,000 and go from there.

I am looking mainly at functional but well-loved equipment on ebay and craigslist, and even some homemade or hacked projects capable of carrying out the job at hand. EBay has some amazing deals. So does Amazon (I usually avoid Thermo subsidiaries as much as I avoid Wally World!). And if you are willing to do some soldering, coding, and 3D printing, you can make just about anything in the lab. I acknowledge that reliability will undoubtedly be less than brand new equipment still on warranty, but of the three-legged stool of quality, cost, and time- this exercise is focused primarily on cost.

Here is my initial list of equipment, projected cost, and notes on procurement. Let me know what you think, where you think I can cut the cost down, or if there are any glaring errors in my choices or logic. I plan to revise and post updates to my budget biotech quest, so hopefully this is only the beginning!


Item Cost Notes
pcr – thermal cycler $200.00 used… Ebay. Or ghetto fab water bath and servo setup?
gel rig $53.00 kit from
light box $80.00 mini kit from or build myself
pipetters $100.00 2 pipettes at 50 each. P20 and P200 to start
water baths $12.99 crock pots with arduino thermostat (DS18B20 and SSR)
-20C freezer / fridge $50.00 total budget – cheap chest – or used on CL
incubator $50.00 Cooler/chest freezer with hairdryer/heatgun heater and extra fan? Or foam insulation box.
autoclave $83.99 pressure cooker- presto on amazon. Ikea?
centrifuge $80.00 used Brinkman on Ebay
flame for plate work $60.00 Ebay flame/regulator kit
Micro scale $25.00 amazon
microwave? $20.00 craigslist
gel power supply $15.00 homebuilt (budget molecular biology power supply)
rotary shaker $140.00 used old NewBrunswick on ebay
pH meter $20.00 Amazon or Ebay
Item Cost Notes
tips $10.00 Ebay- bulk tips – need upfront tip box
weigh boats $10.00 or papers
DNA stain $68.00 SYBR Safe – cheaper alternatives with blue LED illuminator?
gel buffer TAE homemade
DNA polymerase $39.00 mastermind from
dNTPs included in mastermind mix
DNA buffer TAE homemade
gel extraction kit NA45 paper? Simple protocol? Kit…
1kb ladder $89.00
load dye bromophenol blue
comp cells homemade – but lack of -80 will affect competency
Total Consumables: $216.00
Total Combined: $1,205.98 over budget already! And chemicals not even included yet!

Builds I’ll get to eventually

So I’m swamped with work right now, but I have a number of projects that I want to write up and share. For now I’ll just show some of my old pictures and give a quick overview, and hopefully it motivates me to finish everything. Hopefully…

Mini bioreactor controller
Arduino controlled peristaltic pumps via relays with 20x4 LCD
I tried making a mini bioreactor unit to support smaller spinflasks with feeding and to operate the mini flowcell spectrophotometer, and got it to work decently well. There were two peristaltic pumps from Adafruit, a 20×4 LCD to show the time and culture status info, and some buttons for manual advance of the pumps.

Temp display and datalogger
High temperature display and SD datalogger
To monitor vessel temperature while running processes, I used a DS18B20 digital temp sensor with my perma/proto Uno. I now use Arduino pro mini’s for my embedded projects, but back then I didn’t know what was up.

Millisecond light timer switch
arduino, ds1307 RTC module, and relays to switch things quickly
I did a project that required turning a light on for one second every 24 hours. Conventional analog light timers give you half hour resolution, and some digital light timers can get down to the minute. We needed seconds, so I turned to an arduino with a DS1307 real time clock module to keep time, and relays to handle the heavy lifting.

Light sensor and datalogger
photoresistors via arduino to micro sd datalog
To make sure I was switching the lights at the correct time and duration, I used a different arduino to measure the light levels and datalog them to a micro SD card on an ethernet shield. Photoresistors, arduino, sd card. Boom.

Bioreactor datalogging and sending to the cloud
BioFlo 3000 arduino connection
Old bioreactors are cheap and capable, but they’re not very digitally advanced. But the old data access ports can be tapped into for datalogging and triggering alarms through an arduino. This is a custom cable with Cat5 cable soldered to a 9 pin din connector to a New Brunswick BioFlo 3000 bioreactor. Different iterations had different capabilities, but pulling the data off the machine and either datalogging, or uploading the data to an M2X server, or email alerts were used for different experiments.

3D printed protein crystal structure models
3D protein model crystal structure
I know it has been done before, but I had to include a picture of my 3D printed protein model. They are just too cool!

I’ll try to sit down one of these days and start to write things up, but if there is something you want more information about, email me or leave a message in the comments section!


Flow cell spectrophotometer

Mini flowcell spec

While building a mini bioreactor I thought it would be cool to have a personal spectrophotometer to monitor the culture density over the life of the culture. I was inspired when I saw a product from Ibidi for doing live microscopy in an aseptic, optical grade flow cell with luer lock fittings on the inlet and outlet. Boom! A disposable flow cell solution for my mini spectrophotometer system. They are offered in different flow chamber thicknesses and even surface coatings for adherent cell growth, but for a yeast or bacterial culture I opted for the untreated or non-functionalized chamber.

PCB board with LED and digital light detector
Link to full size image

On a scrap pcb I rigged a light to frequency converter from with a 0.1uF capacitor, powered by the 5v from an Arduino, and for the light source I used a LED that emits about 600nm light, and to cut down on background light I even cannabilized a floatie to act as the light chamber (I may have also won an award for it here). The code is pretty simple, turn on the LED, pump the culture through sterile peristaltic pump tubing and through the flowcell (pump activation time depends on the length and volume of tubing), measure the amount of light that is transmitted through the culture, compare the reading to known values, and pump the culture back into the spinflask or bioreactor. As nanodrop spectrophotometers exemplify, you don’t have to have a big cuvette with 1cm wavelength to take an A595 reading. Beer’s/Lambert’s law is dependent on the pathlength, so the math in the code depends on the chamber thickness of the ibidi slide. Or it is somewhat irrelevant because it will all be calibrated with known culture concentrations that have been measured with a true spectrophotometer.

Peristaltic pumps are pretty easy to come by (adafruit, ebay, etc). I love the availability and functionality of pharmacia P-series pumps, but I can’t get the full functionality that is suggested in the spec sheets. Through a 15pin din connector you should have the ability to engage the motor, change the flowrate, and even change the directionality of the pump head, but admittedly I can only get an arduino to turn them on and off. I need some more time to tinker with the communication, but I suspect it might require 5-20mA communication, as that seems to be the industry standard for PLC controlled instruments. On the open end of the peristaltic pump tubing beyond the slide/flowcell, I put a hydrophobic 0.45 micron sterile filter. The tubing can all be autoclaved with the reactor vessel, and then aseptically put on a sterile ibidi flowcell and place into the LED/sensor assembly.

I’ll dig through my backup drives to find the code, but I must have archived it a while back. For what it’s worth, I used the basic DS1307 real time clock code to keep time, used the TSL235R example code to pull the signal, sampled on a set interval, and then datalogged the values onto a micro SD card through an ethernet shield on an Arduino Uno. I was happy with how it turned out, and that GoldBio thought my alternative use of their floatie was pretty cool!