Beyond the laboratory notebook: Why project infrastructure quality matters in your PhD/research career

Over the last nine months, I have been given the opportunity to train fellow colleagues within my research team about the very unique experimental methodologies I’d developed as part of my research project. I’ll be clear: I didn’t really need to do this as part of my PhD candidature, and I know that some people will put forward the point that it could be seen as procrastination, or a distraction from other examinable/grant-pulling aspects of my PhD.

However, I’d like to showcase the reasoning why I decided to take on an extra burden of investing nine months for training staff with an example, as fundamentally, it is a time investment that pays dividends in the form of clearer thinking, a higher quality thesis, regulatory compliance and a clear path to project successionship.

A common story that I hear from a lot of young researchers in the academic research arena is that when they take up a research project, they discover that the previous student kept an absolutely horrendous set of notes in their laboratory notebook, where protocols and experiments are poorly documented, and instructions on how to operate and calibrate specialist equipment are missing. For whatever reasons, Good Laboratory Practices (GLP) aren’t really being adhered to. The student then discovers that for the experiments to work, a larger set of resources and infrastructure (technicians and other project stakeholders) are required to really make things happen, beyond the standard laboratory consumables, equipment, etc.

Generally, institutions where the research takes place will implement some form of laboratory notebook control to ensure that good laboratory practices are being adhered to. Certainly, the level of compliance varies from place to place, but overall, the need for IP protection is a strong motivator for a high-level of compliance.

What is less commonly discussed is the documentation and procedures put around the entire project. I’m an engineer, so I see the world as systems, interfaces and interactions. I like to consider scientific experiments as a component of the much larger system that is designed to “feed in” a question and ‘”spit out'” a research finding and IP. These systems are unique to each individual research project, however, the fundamentals about how they are set up and managed are much the same.

My situation 9 months ago looked like this: I needed further ethical approval to conduct studies, and a reshuffling of staff within our team had meant that I had a research assistant available to me to assist with the very labour intensive project I was developing. I had been troubleshooting the technical methods of my project for at least 18 months, but the bugs were starting to be ironed out so that I would be able to run a series of experiments under fixed technical conditions. It was also becoming clear to me that I needed a second pair of hands – up to this point my PhD had been run quite independently by yours truly, but I was acutely aware that the research tasks were become evermore time consuming and overall, would progress beyond my candidature, and that the team would need to continue everything that I was doing.

I had two options: The first was to continue running experiments independent of the research team, and effectively “give up” the “helping hand” I had available to me. I was confident in the execution of my experiments (bugs and all) and understood what to do when something didn’t work out (in order to attempt a salvage any wayward experiments). However, I still required ethical approval for experiments beyond specific milestones, and continuing the project at the pace I was driving it would have landed me with a lot of heavy-duty, concurrent tasks. A recipe for stress and PhD-induced depression.

I needed a better alternative, so I used a bit of outcome-driven thinking to come up with an alternative solution. I considered what I needed to achieve for my PhD and my team collectively:

  1. Complete, unified ethical approval for the project. Up to this point, the ethical approvals were disparate – many components of my project had been considered by the ethics committee as different sub-projects and pilot studies, causing severe administrative difficulties. In the short term, I, and in the long term, my team, required a single, unifying project application that gave us approval for all components of the project
  2. A complete replication of my skills and technical know-how for the project. I needed the staff I worked with to completely understand how my project worked and how to run it, but most importantly, what to do when things didn’t work. I needed an independent-minded research assistant, as opposed to a programmed robot/trained primate.
  3. A data storage structure that everyone involved in my project could understand, so that my team could access my data and analyses.

The solution seemed counter-intuitive everything that was being encouraged of me to complete my PhD.

I needed to slow down my project schedule in order to spend time preparing for it to be scaled up.

Yes, I can hear you gasp. But hear me out.

In order to meet the first 2 requirements, it entailed the development a series of Standardised Work Instructions (SWI) that brought together all the methods and protocols into a unified format. One protocol, one document – describing all aspects of the work that needed to be completed; apparatus, resources, people and time scheduling issues. These SWIs would form the basis of a training platform and be the centrepiece of my future ethics applications. They would allow me to a) train the research assistant to the required capabilities (both in technical skill and project management) and b) create a new ethics project without complex detail “choking” the flow of the application document.

Writing out the SWIs would take time and involve a large number of stakeholders. I had to factor these elements in, hence slowing down certain processes within my existing project workflow. It took me close to 3 months to write the SWIs, and a further 6 months to gain ethical approval. Over that time, I prepared for workflow changeover and incrementally trained staff to replicate the things I did to a high degree of detail, as attention to detail was critical to project success. I’ll be honest, it wasn’t easy. It tested my patience to the limit, and then broke it. I’d previously trained fellow colleagues when I worked in photographic retail as an undergraduate, but this was a big step above.

To address the third requirement, I had to overcome every researcher’s fear of opening up their data to the world. Up to a point, I had all my data, data analysis and associated files all kept on hard drives that only I had access to. I realised that in order for my project to move forward, data transparency and accessibility within the team had to be established. It meant breaking up my personal filing system into separate parts, so that the key components of the data I’d accumulated could be accessed centrally. What helped this process was telling myself that the PhD was still mine, but the data were not. The data were my group’s, not mine. Once I convinced myself of this, I was able to break that emotional bond between myself and the data. As I was creating the filing system, I also took the opportunity to ensure that the data could be easily found and accessed, complete with documentation on where specific things were kept.

So it’s clear that I spent a lot of time and effort on setting up quality infrastructure around my project. But I think the time invested will yield much higher returns, such as:

  1. The ability to now truly focus on writing my thesis, while my team members continue to lead the project independently
  2. Using the SWIs as part of any methods I describe in my thesis and publications produced either by me or in future by my team, and know that they are of a high standard
  3. Enabling my supervisors to better supervise me, as they now have access to all of the important information they need to help me complete my candidature on time
  4. The ability to train even more staff in future, maybe even find the project’s scientific successor.

One unexpected positive from this entire process was that we were able to show we were compliant with government regulations. Aspects of our work were externally audited shortly after I completed the SWIs. I was terrified, I couldn’t afford to make any mistakes, especially as it was an external audit. My foresight, however, paid off. Through good record-keeping and having the SWIs complete, we were able to show the auditors that not only were we doing good research, but translating our methods into documented, high-standard practice. It was fantastic validation that we were on the right path; the audit process for our team was straightforward.

As far as the thesis and examination is concerned, these activities bear no impact on the overall grade. In the current model of the PhD, you’re not assessed on these skills and capabilities that you pick up along the way. However, as an achievement, it’s probably going to be one of the bigger ones I’ll be able to add to the CV:

Designed, implemented and managed a small R&D program. Trained staff in unique, research-specific technical methods. Enabled team members to lead the project.

Often I hear PhD graduates state that they don’t have the skills that industry are looking for. My personal belief is that within your PhD, you do have the flexibility to create your own opportunities such as the one I’ve described above, and if you can align them with the skills that are valued outside of academic research, you are setting yourself up for success. And of course, your team will also reap the benefits! That new grad student? They’ll be able to walk in and pick up the important things they need to know quickly and easily, and at the same time they’re learning about good laboratory and research practice. That paper your supervisor/PI/collaborator needs data for? Easy – just tell them the location and they can find it. Easier data access = easier manuscript preparation = faster submissions. So it’s certainly in your interest to invest time in setting up your project infrastructure properly and to a high standard.

Until the end of my candidature, this project’s management and leadership will need further fine-tuning, but I can now do this from a distance. I’m now able to dedicate myself to the task of writing, knowing that my team has the skills, capability and flexibility to continue the work I’ve developed. Whether you’re a grad student, early-career researcher or are established, I hope this post inspires you to create high-quality research project infrastructures – the benefits are too hard to turn down.



One thought on “Beyond the laboratory notebook: Why project infrastructure quality matters in your PhD/research career

  1. In my thesis lab we would always talk about how “You may think you never have time to do it right, but then you will find the time to do it twice.” I agree time spent slowing, down and getting organized to “do it right” pays off in the long run. Project management skills and proper delegation are one of the many unsung soft skills of PhD holders.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s