https://images.pexels.com/photos/48604/pexels-photo-48604.jpeg
You can replace a laptop. You can’t always replace the only record of your experiments over the last 6 months.
When running a research lab there are a few additional issues you need to think about. Perhaps a sample you have used up, equipment that you will never get your hands on again or results you may struggle to replicate.
Ensuring that you have good cyber security can help protect research records, allow you to track results and continue your work in case of a failure.
What you’ll learn:
- Start With the Records You Cannot Replace
- Keep Sample Details Connected to Results
- Protect Access Without Slowing Down the Lab
- Build Backups That Can Actually Be Restored
- Know What to Do When Something Goes Wrong
- Wrapping It Up
Start With the Records You Cannot Replace
You are not going to want to secure everything to the highest degree. You can always recreate some documents such as minutes from a meeting. However you may not be able to recreate some of your raw data from an instrument if the experiment only occurred once.
Try and list the data you will find most difficult to recreate. This could be lab notebooks, raw data, instrument settings, calibration information, scripts used to analyse your data and sample information.
Ask yourself the following questions about these datasets:
- Where is the original record of this data located?
- Who is looking after this data?
- Who can alter or delete this data?
- How soon would you need to recover this data if it went missing?
Create a simple data register to hold this information. Assign a person, location and importance level to these key pieces of data. This will identify where you may be holding important records on an instrument PC or laptop.
One question you also need to consider is how long you need to keep these records for. Depending on your institution’s, funders or project requirements you may need to retain your data for a certain period. Don’t clear old data from your instrument just because it will free up space. If you need to reference those results in the future, it can be tricky to retrieve.
Keep Sample Details Connected to Results
It’s all very well storing a result, but being able to identify what you got that result from is also important.
For example if a lab was keeping track of a retatrutide peptide sample they could store a reference to their product retatrutide 10mg. From here they should be able to link that product reference to their sample, lot number (if applicable) and what work was completed using that sample.
Make sure you have a record that can track the journey of your sample from start to finish. Note down your sample ID, information about when it was received, any important information about your sample, what experiment ID you used and where your raw data is stored.
Ensure that you use the same sample ID throughout your notebooks, exported instrument data and analysis files. Having one typo in your file name can cause you to lose track of where a result came from.
The NIH advises to have metadata available about your research data, including things like methods, labels and definitions of variables. By doing this you can ensure that your data can be used by others in the future.
Ensure that you keep your raw data files as they are. If you need to modify the data or update any analysis scripts, create a new version and note what you have changed. This will allow you to double check your results and look at older versions of your analysis if needed.
Protect Access Without Slowing Down the Lab
Sharing logins across multiple users can be handy but can also become an issue. If you ever notice your data being modified, it may be difficult to tell which person changed the file.
Allow your researchers to have their own login where possible. Grant them access to the data they require for their job role and keep admin privileges limited to those who need it.
There have been many cases of researchers falling victim to cyber security issues. According to the UK’s Cyber Security Breaches Survey for 2025/26, 98% of higher education institutions identified a breach or attack in the 12 months.
Ensure that your emails, cloud drives and remote logins have two factor authentication if you can. Always check if anyone needs their access revoked when they move to a new position or leave the organisation.
Some instruments can be difficult to secure as they may run outdated software that cannot be updated without losing functionality on your instrument. Speak to your IT department to see if these can be isolated and only connected to certain devices.
Remember to keep an eye on any data you share with collaborators. They may need access to one specific project, not all the folders on your lab drive. Share through approved methods and double check access and when that access should be removed.
Have logs of any access or changes made to important data where possible. This will help your IT/security team understand what has happened if they need to look into any suspicious activity.
Build Backups That Can Actually Be Restored
Having a copy of your final spreadsheet is better than nothing but won’t help you recreate your experiment.
Try and backup as much information as you can, including your raw data, sample information, notebooks, scripts and any files that can read your data. Make sure you note any additional software or licences you may need in order to access your data.
Don’t fall into the trap of thinking you have backed up your data by syncing your files to multiple locations. If that file gets corrupted or deleted, it could sync to all your other devices.
As the UK’s NCSC suggests, ensure your backup is secure and away from your main device and that you can recover it. Make sure you have a plan in place to:
- Automatically save important files, keeping previous versions.
- Have a secure location that cannot be accessed by default users to delete or alter your backup.
- Recover a set of files to a test location and ensure you can open them.
- Ensure any scripts, settings and sample information are also able to be accessed.
Practice this with your own type of data, you may be able to open your recovered documents but struggle to access files from an instrument or run an analysis script.
Record any problems you come across when testing and resolve them before you need to use your backup in an emergency.
Know What to Do When Something Goes Wrong
Despite having a good backup in place, you will still want to know what to do if data is altered, held to ransom or you can’t access a system.
Have someone in mind that your researchers can contact if they notice anything unusual happening with their logins, data or instruments. Let them know that their first port of call is to report the issue, not delete anything they think looks suspicious.
You may want to halt some work if you can’t trust your records to be saved properly. Consider if you have any work that would need to be stopped, perhaps if you are currently collecting data from an instrument you would want to halt this but may be able to continue if it’s an old dataset you are working on.
Keep contact information, a list of your systems and how to access your backups somewhere that you can reach if your main system is offline. Someone will need to be able to contact your institution’s IT/security team without needing to use your main email account.
After you have regained access to your systems, ensure your data is correct before you begin working again. Identify what data is safe, if any of your records have been tampered with and what you need to do to reconcile any work completed while you had no access.
Practise with your team, pretend that your main drive has failed for your current project. Do you know who is responsible, what work you need to halt and where your backup is?
If you don’t, update your plan and try again.
Wrapping It Up
The data you care about most in your lab is probably something you can’t recreate.
Know what data you need to secure, keep track of your samples, restrict access to those who need it and test your backup regularly.
