Bringing Clarity to a Complex Proteomics Study through Bioinformatics Mentoring

When PhD student Kirsikka Musta and her colleagues at Tampere University, Finland, generated both human and bacterial proteomics data from coronary artery samples, they found themselves working with an unusually complex study design and no obvious data analysis workflow to follow. To move the project forward with confidence, the team turned to Genevia Technologies for expert bioinformatics support.

Building bioinformatics know-how through specialist support

At Tampere University, the Clinical Chemistry group, led by Professor Terho Lehtimäki, investigates various aspects of cardiovascular disease, ranging from vascular and cell biology to genetic, epigenetic, and biochemical risk factors and atherosclerosis biomarkers. As part of this work, PhD student and molecular biologist Kirsikka Musta was studying coronary artery samples using mass spectrometry-based proteomics. The project soon raised an unusual data analysis challenge, exceeding the group’s internal bioinformatics capabilities: the dataset included both human and bacterial proteomes from the same samples.

Understanding the analysis, not just running it

Genevia’s senior bioinformatician, Dr. Efstathios Vlachavas, supported the project with a practical and pedagogical approach. After reviewing a sample of the data, he prepared a template R script that walked through the key steps of the analysis workflow, with clear explanations and comments throughout.

“He provided me with a script template and added questions and comments, such as ‘think about why this is done like this,’” Kirsikka says. “It was a great pedagogical approach for teaching bioinformatics.”

The work progressed iteratively. Kirsikka would apply a specific part of the analysis pipeline to her data, make observations, gather questions, and then discuss them with Efstathios in a meeting. Much of the communication took place by email, supported by regular video calls.

“It was really good to have the human proteome script as a point of reference and as a template for the full workflow,” she says. “It helped me understand the entire process.”

One of the most valuable parts of the collaboration was the depth of discussion around methodological choices. Before working with Genevia, the team had spent considerable time trying to understand imputation in mass spectrometry data: when it is needed, which methods are appropriate, and how different strategies affect the results.

Together with Efstathios, Kirsikka compared different imputation approaches in detail and developed a much clearer understanding of when each method should be used.

“It was great that we could go that deep into it, rather than blindly selecting a method and being left with an insecure feeling about whether it was the right choice,” Kirsikka says.

The collaboration also helped her learn how to perform differential abundance analysis using an R package called limma, and, importantly, understand what was happening behind the scenes.

“I didn’t just learn how to run the analysis. I got to take a look under the hood.”

The scripts became the main deliverables of the project, but the value extended beyond the code itself. The collaboration helped the team understand conventions and best practices for handling proteomics data in R, giving them knowledge they can apply in future projects.

Bioinformatics support for a non-traditional study design

The biological setup of the project was highly multidisciplinary. Coronary artery samples are typically available only through clinical collaborations, and including bacterial proteome data alongside human proteomics made the study especially unusual.

Kirsikka was impressed by how well Efstathios understood the biological research questions and contributed ideas for approaching the data analysis challenge.

“There is no gold standard for how to deal with this kind of study design,” she says. “But he had so many great ideas for different bioinformatics strategies we could apply to our dataset.”

This combination of computational expertise and biological understanding helped the team move faster and with more confidence.

“He was a massive help,” Kirsikka says. “We could not have done all this if it wasn’t for him, or at least we would not have been able to progress as quickly by ourselves.”

Saving months of work through flexible, clear, and confidence-building collaboration

For Kirsikka, the collaboration stood out not only because of the technical expertise, but also because of the clarity of communication. Efstathios kept her well informed about the hours used, provided realistic estimates for different parts of the work, and answered questions in a structured and detailed way.

“I especially liked how clearly and in detail he provided answers to all my questions,” Kirsikka says. “He formatted the emails very clearly, dividing the replies into sections, which helped me a lot to follow and understand.”

This was particularly valuable because English is not Kirsikka’s first language and the project involved many specialized concepts.

The flexibility of the collaboration was also important. The team had considered alternatives, such as attending a summer school to learn proteomics data analysis, but that would have meant delays, travel, and uncertainty about whether the course would provide enough project-specific support. Consultation with Genevia proved to be faster, more practical, and more cost-effective.

“We started working with Genevia early this year. It’s early June now, and we’re finalizing the project,” Kirsikka says. “If we had waited for a summer school, we would not have even started yet.”

Looking back, Kirsikka and her supervisors see the consultation as a way to accelerate both the analysis and the learning process. Without close colleagues available to support this specific type of bioinformatics work, learning everything from scratch would have required substantial time and effort.

“If I had started to learn all of this on my own, the working hours and salaries required would have become a lot more expensive than buying the consultation,” she says. “In this kind of academic setting, we are definitely talking about months of working time saved.”

The collaboration also gave the team confidence that the analyses were carried out according to best practices, potentially helping them avoid time-consuming revision rounds later in the publication process.

“When you do things like this on your own, there are always some details that go unnoticed, and fixing those takes a lot of time and effort,” Kirsikka says. “This consultation gave us confidence that everything was done according to best practices.”

A new perspective on coronary artery disease

The bioinformatics part of the project is now close to completion, and the team is continuing with additional imaging work to support the findings. The results have already provided intriguing insights into coronary artery disease.

“We are very happy that this project and collaboration helped us gain this new perspective to coronary artery disease,” Kirsikka says. “Whatever the final results will look like, we already know that this project will improve understanding of the disease.”

For future projects, Kirsikka says she would readily consider working with Genevia again, especially when specialized bioinformatics expertise is not available within her immediate research network.

“The field is moving extremely fast, and the need to keep learning is never going to go away,” she says. “I would definitely consider working with Genevia again.”

Her message to the Genevia team is simple:

“Great job!”

Read more

• Bioinformatics Mentor
• Proteomic and metabolomic data analysis

Contact us

Leave us a message below if you would like to hear how we can help you.