Our bioinformatics training programme has proved hugely popular since being launched in 2022. Delivered in collaboration with the Glasgow Bioinformatic Core, these courses equip wet-lab immunologists, biologists and other life scientists with the skills and confidence to perform their own bioinformatic data analysis. Here, course participant, Dr Olivia Bracken, shares her experience of taking the first two courses in the programme.
Having always been a purely wet-lab scientist, I was apprehensive to enter the world of bioinformatics. However, the explosion in the use of large-scale datasets in scientific discovery meant that I knew I needed to take the leap and delve into the world of R coding, if I wasn’t to be left behind.
Data wrangling
I initially signed up for the BSI’s beginner course ‘Omic data analysis and visualisation in R’, which gave an excellent grounding in data wrangling and presentation of omic data. John Cole seamlessly took us through how to manipulate our data into the correct format to create the multitude of different plots that can be generated in R, as well as how to customise these. Not only did John make learning the language accessible, but the lectures brilliantly outlined the importance of different plots in presenting various datasets, and how these can be used when preparing data for publication. I was hooked and immediately set about applying what I had learnt to proteomic and RNASeq datasets that we had available in the lab.
When the opportunity arose to take the second course, ‘Further omics, statistics and clinical data in R’, I jumped at the chance. This course makes use of two datasets – proteomic and RNAseq – to interrogate metadata, code your own functions and apply ‘clustering’ to assign different cell types. The proteomic dataset came with accompanying clinical metadata, which enabled us to learn how to look for correlations and perform statistical analysis. Perhaps most helpful (and interesting) was how to make our own functions so that we could streamline our analysis. I thoroughly enjoyed applying strategic thinking to building a function, and John and his team were on hand to guide us through the steps and answer any questions that arose.
One step at a time
Next, we used an RNASeq dataset to work through the DESeq2 package to analyse bulk RNA sequencing. Again, the combination of the lectures and the two-hour tutorials gave time for John to explain in detail how RNAseq experiments are prepared, the importance of batch correction, and when to apply quality control corrections, all of which meant that when we came to applying the code to the dataset, we had a clear understanding of why each step was being performed and how it would affect our result. The additional analysis techniques we learnt were brilliant – we went from being able to compare two conditions to comparing three.
I certainly feel that this latest course has taken my coding skills to the next level. I can interrogate datasets in a way I wasn’t able to before. John is such an excellent teacher – he truly makes the whole experience so enjoyable. No question feels too small and, even after completing the course, you can always drop him an email should you have any further questions. Having completed these courses, I feel I am capable of navigating any package in R. I have analysed RNA datasets using DESeq, proteomics with MS-DAP and flow cytometry data with CATALYST. The second course gave me the chance to return to these packages newly equipped to manipulate the existing code to fit my datasets.
I never thought that I would become a coder. I was certain I would remain a wet-lab scientist forever, but completing the R courses with John has completely changed my mind. John has built courses that mean no problem feels insurmountable. He makes it accessible, interesting and I couldn’t recommend the course highly enough. I now love coding, and am just sad I didn’t pick up this skill sooner. I feel I am a much better scientist for it.
Dr Olivia Bracken, University College London
Find out more about all our training courses here.