MSci Neuroscience

Computational methods are increasingly used in all areas of biology in order to better understand complex living systems and develop models that generate testable predictions. This unit introduces a range of computational techniques, including differential equations, machine-learning, network and constraint-based analyses, which are used for a wide range of biomedical and biotechnological applications, from understanding how intracellular signalling pathways are disrupted in diseases to redesigning organisms by metabolic engineering.

A-Level Mathematics required.

A limited number of places will be available for BSc students, the following rules apply:

1. The student MUST have A’Level Maths (or its international equivalent - but AS or GCSE Maths are not sufficient).
2. The student cannot join the unit in the 2-week grace period at the start of the semester, they would have to be pre-accepted in early September.
3. The student must first discuss with their PD whether taking the unit is appropriate for them or not.
4. Where demand exceeds capacity, acceptance onto the unit will be based on the student’s Year-2 average mark.

This unit aims to introduce students to a wide range of computational methods and tools required to carry out interdisciplinary research in the biological sciences.

The unit will start with an introduction to the Python programming language. Students will be introduced to the Jupyter Notebook system, a widely used online application allowing the development of code for data analysis and numerical simulation. The core of the unit will be structured along four main sections, each covering a particular set of techniques and applications:  

Section 1: Dimensionality reduction and clustering 

• Dimensionality reduction: Principal Component Analysis (PCA) and interactive plotting with applications to visualising single-cell expression data 

• Clustering: hierarchical, k-means and mixture model clustering 

• Non-linear dimensionality reduction methods: GPLVM and t-SNE for non-linear dimensionality reduction and visualisation of single-cell data  

Section 2: Models of large cellular systems 

• Network reconstruction and analysis: protein-protein interaction networks, metrics for network analysis, integration of high-throughput biological data

• Logical modelling: Boolean models, logical steady state analysis, applications to cancer systems

• Constraint-based modelling of metabolic networks

Section 3: Dynamic models 

• Introduction to differential equations-based modelling 

• Modelling gene regulatory pathways: gene transcription and cellular signalling pathways  

Section 4: Next-generation sequencing  

• Next-generation sequencing I: from laboratory experiments to computational analysis

• Next-generation sequencing II: RNA-seq workflow

The unit will be delivered as a succession of lectures (1 hour) and practical sessions (2 hours), where each lecture will introduce the theory behind a method/tool, and students will apply the method/tool to solve a particular biological problem in the practical. Students will be assessed by completing three written assessments, one for each of the main sections of the unit. These assessments will consist of a series of short questions and mini-project reports, some of which will require some computer code to be written.

Students should: 
• Understand essential mathematical concepts required for biological research. 
• Understand and apply differential equations modelling of intracellular systems. 
• Understand and apply constraint-based modelling of metabolic systems. 
• Understand and apply network analysis and logical modelling of molecular systems. 
• Understand and apply data analysis and machine learning methods. 
• Understand the applications and limitations of different modelling techniques and tools. 
• Understand and use RNA-sequencing data analysis methods.

Students should:

• Develop problem-solving skills.
• Construct models and design experiments to test biological hypotheses.

Students should:

• Use the Python language and develop models using the Jupyter Notebook system.
• Construct models of signalling, regulatory and metabolic systems.
• Infer computational models from biological data.

Students should: 

• Develop computational skills.
• Develop report writing skills.
• Learn to communicate computational results.

Three online modules worth 30, 40 and 30% respectively.

Verbal feedback will be communicated during the practical sessions.

Written feedback will be communicated through annotated comments for each online assessment.

Specific material will be provided with each lecture.