How to Pick a Picker


What you should consider when automating colony picking

Is colony picking a bottleneck for you?

Colony picking can be performed manually but, as your lab grows, so too does your need for efficiency, accuracy and throughput. This guide describes some of the many automated colony pickers on the market, taking you through the main considerations when choosing a solution right for your workflow.

Not sure whether colony picking is a bottleneck for you? Our prioritising bottlenecks guide may help.

Colony picks per hour: speed bumps

The throughput of colony pickers, often quoted in picks per hour, is often the headline figure that manufacturers present about their products and it can give the impression that this is the most important consideration about colony pickers. It shouldn’t be.

Some manufacturers claim to pick up to an impressive 3000 colonies per hour. However, these numbers are achieved under optimal conditions and don’t always factor in important workflow steps (such as sterilisation, mixing, optimising speed to reduce contamination). The advertised picking rate also does not take into consideration all the other factors that reduce the real picking rate.  Downtime, reliability, mechanical accuracy can all eat into this headline figure for throughput . It is therefore important to take a holistic view of the features on offer. In the following sections, each of these potential bottlenecks is explored in more detail.

Consistency is key to reproducibility

Up to 50% of scientists have failed to reproduce their own experiments and experimental error is a major source of problems in reproducibility (Baker, 2016, Nature). So even though some manufacturers like to boast about their picking speeds, we think precision and accuracy are far more important. Check out why reliability beats speed in colony picking to find out more.

Types of error in colony picking include missed pinning events which reduce picking efficiency and cross contamination (picking the wrong colony).

In colony picking, picking error must be factored in during experimental design by increasing the number of replicates. Not only does having to increase the number of replicates increase the use of resources, it also eats into the effective picking rate. It is therefore advantageous to minimise cross contamination and maximise picking efficiency. Faster systems frequently cause cross contamination as the impact of the pinning surface against the agar can cause small amounts of ‘splashing’. It only takes one cell from a colony to become aerosolised which can then land somewhere else on the plate leading to cross contamination. Optimising the ‘kinetic profile’ of the pinning heading can overcome this but this can come at a price of picks per hour. Picking efficiency is often reported to be between 95% to 99% but methods of measuring this differ between manufacturers. When speaking with manufacturers, ask them to verify the claims they make about picking efficiency and whether the reported efficiency can be achieved at the throughput advertised.

Accuracy during pining to agar is also critical because this medium is quite fragile. Agar can be accidentally pierced by the pinning surface at pressures as low as 12 g/mm2.  Agar is frequently damaged by some systems, particularly at higher speeds. This may be acceptable but biological material accidentally pushed into the agar may not grow properly and may affect downstream processes such as subsequent pinning or analysis of colony growth. When evaluating a platform, make sure to check if the instrument is capable of accurately determining the surface of the agar to minimise the risk of agar damage.

Pinning and sterilisation: clean it or bin it?

A colony picker is meant to move biological material from position A to position B. But as with any technique is microbiology, a colony picker can also end up moving contamination as well. There are two sources of potential contamination during colony picking: environmental (e.g. from the air) and from other samples (cross contamination). Colony pickers employ a wide range of features to minimise contamination.

Most colony pickers use metal pins that either need to be switched out [to be cleaned externally] or automatically cleaned. Automated cleaning can use baths of ethanol and/or bleach to clean the pins ready to pick again. This technique can work well but contamination is possible when picking mucoid- or biofilm forming microorganisms which can resist the cleaning step. Residual ethanol and bleach left on the surface of the pins can kill samples including E. coli and algae. These baths need daily maintenance adding to on-going costs and reducing walk-away time. Some instruments heat the metal pins to sterilise. This is effective but is a known bottleneck in some systems (which may not be quoted in the headline picking speed). 

Automatic colony pickers that use a polymer filament avoid these issues by cutting off and disposing of the small amount of filament that has been in contact with the sample which reveals fresh clean filament for the next pick. The reels of filament can last for many thousands of picks which minimises maintenance and generates little waste. The filament should be sterilised during manufacture and heat treated as it ejected from the pinning nozzle to ensure it remains sterile.

To minimise environmental contamination, some (but not all) instruments are sold in enclosures. Instruments without enclosures can jeopardise the sterility of samples and putting these instruments in enclosures can come at a considerable cost. Another advantage of some enclosed systems is that they can have the added feature of UV sterilisation. UV is a handy feature for keeping your instrument clean and minimising contamination, and can be an added safety feature when working with hazardous microorganisms.

Reliability: Minimising downtime and maximising walk-away time

Picking speed is irrelevant if your picker isn’t picking. We know from managers of large automated laboratories, reliability is often the single biggest bottleneck. Manufacturers are not always upfront about the reliability of their instruments but it is a fact, all mechanical systems will have downtime at some time or another. Due to the lack of transparency, it therefore makes it difficult to compare different systems. Read customer reviews and testimonials to find out users’ biggest gripes are with specific technologies. If customers are reporting frequent issues, this can point to serious reliability problems or customer support issues that will create a lot of downtime and ultimately become a bottleneck in your workflow. One way to minimise downtime is to perform maintenance tasks in a timely manner and have your instruments serviced at the recommended frequency. Look for instruments that allow the user to perform basic maintenance as this reduces the need for costly visits from an engineer; saving time and money.

Versatility: What else do you want in your colony picker?

The standard workflow that all colony pickers can do (to a lesser or greater extent!) is pick colony material from one plate and move it to another. But how else can a colony picker be used and what are the features to look out for?

Extensibility is a great feature to look out for in a colony picker. Instruments that have an open API can be integrated into larger automated systems including incubators, plate readers and robotic arms. If you’re comfortable with programming, the API can also be used to create custom workflows that can’t be achieved using the standard software. If you don’t feel comfortable programming in code, some platforms have the option to program the instrument in a more human readable format such as importing a list of commands (e.g. Pick from A1 in Plate 1 to A1 in Plate 2).

Does your colony picker need to go in an Anaerobic chamber? If so, there are some important considerations. Some instruments are very large and creating custom chambers to accommodate these instruments is very expensive. Larger chambers require more anaerobic gas mix and are more expensive to run. Keeping the instrument on the smaller side can keep costs lower. Some instruments run on compressed gas which can interfere with the balance of gases in the chamber. This in turn requires more anaerobic gas mix to be used to purge the system or can require more hefty oxygen scrubbing to keep the conditions right.

Specialist workflows may require the ability to screen for fluorescent colonies, or pick to specialist target plates for MALDI-TOF microbial identification. Some colony pickers have been developed to be able to pick colonies from inside zones of inhibition. Talk with the manufacturer to find out if they offer these features and to what extent these features have been validated.

Ease of use: Don’t get crushed by the white elephant

Many robotic systems are complicated to operate and maintain which reduces the usefulness of the instrument and makes it more difficult to train people to a competent level. Those working in laboratories will be all too familiar with ‘white elephants’; pieces of equipment that few [or no one] knows how to operate but are too expensive to update or dispose of. To protect your investment, make sure that the equipment is as simple to use and as easy to maintain as possible. This will maximise the usefulness of the instrument and will make people want to use it. There is no reason why a colony picking robot should be complicated and your equipment will be significantly future-proofed if anyone can learn to use and maintain the instrument with little to no instruction. UX (user experience) should be built into every aspect of the instrument.

Is colony picking a bottleneck in your workflow?

Discover how PIXL can enhance your process with precision, reliability, and versatility.

Phil Kirk PhD | Senior Scientist

Phil heads up our Research team, combining his biological knowledge, programming experience and engineering skills to push our robots to their limits and leading experimental work to back up their use in a variety of applications and settings.

He has a background in plant science and biotechnology, and close to a decade of laboratory experience. Hence, he’s as happy wielding a micropipette as a screwdriver, loves cracking complex scientific problems, and is an absolute whizz at designing an R-script!

By ensuring our innovations do what they are designed to do, Phil’s work contributes to more reliable robots, meaning more time our customers can spend performing the work that can’t be automated: creating, interpreting, and enjoying science!