The taught element of the course gave me knowledge of a variety of areas. There were lectures on proteins, which included protein structure, post-translational modification, visualisation and modelling, as well as methods of determining proteins structure by experiment, such as NMR and X-ray Crystallography. This module also included assessments on protein sequence analysis.

There was also a set of lectures on genes and genomes, which covered genome sequencing techniques (including NGS), transcription and translation, and the enzymes involved, genetic mapping, genome annotation and comparative genomics, and construction of phylogenetic trees. In this module, the assignments covered comparative genomics and the annotation of novel genome sequences.

A number of programming and informatics modules were also a part of the course, with python, java and SQL, as well as a more general algorithms section. There was also a section on statistical analysis using R, and sample bootstrapping to ensure significance. Many of the people on my course had less experience with programming than I did, and I was able to provide help and advice when asked.

In the group project, I was a member of a 4 person group, attempting to replicate and review methods from a gene expression study. We also compared these methods with other, more recent, pipelines with similar applications. The pipelines used were all written in, and accessed through, R, and so provided extensive experience in that language. I also wrote the template for our web page, and used Shiny to generate interactive plots demonstrating our results. The code repository for this project is available on my GitHub page, at https://github.com/MichaelKubiak/Group_Project, and the report can be viewed here. PLEASE NOTE: the webserver on which the website was displayed was private, so cannot be viewed from outside the University of Leicester.

My individual project involved the use of machine learning approaches to annotate protein function from HMMER scores of proteins against the Pfam database. To train these systems, I used protein annotations from Enzyme.dat, the EC annotation file for Swissprot proteins. The manipulation of these large databases provided experience of preprocessing techniques, and required extensive use of SPECTRE, an HPC cluster at University of Leicester. In testing different sets of variables for model training, I had to submit batch jobs to SPECTRE. This project was mainly written in Python, and the code is available on my GitHub, at https://github.com/MichaelKubiak/Annotation. My report for this project is available from here.