This repository is based on my training for Computational Biology at Imperial College London. The scripts and contents find here are based on the TheMulQuaBio repository, authored by Samraat Pawar.
Over this course I will gain competence over programming languages such as BASH, Pyhton and R.
On Week1 we focused on:
-
UNIX and BASH
-
Shell Scripting
-
LaTeX
On Week2 the focus was particularly on Python programming languages. The main learning was based on:
-
Basic syntax and data structures
-
ipython Environment
-
Writing Python code, loops, comprehensions and functions
-
Implementation of control flows
-
Pyhton Input/Output
-
Writing, debbugging, using and testing Python functions
In Week3 we continue to develop skills on Biological computing using R. Particularly, the learning was based on:
-
Basic R syntax and programming conventions
-
Principles of data processing and exploration
-
Clean and efficient programming using R
-
Developing reproducible data analysis workflows
-
Introduction to vectorization and efficiency for simulations
-
Finding, fixing and catching erros with debugging
-
Data wrangling, visualization with tidyverse
-
Introduction to sql databases in R
In Week7 we have turn back again to Python to learn some more numerical computing. Essential the learning was based on:
-
Numerical Integration with numpy and ODE's
-
Random number generation
-
Optimisation adn profiling
-
More vectorisation
-
Regular Expressions
-
Use of subprocess to build computational workflows
In Miniproject folder are the files to run a computational project developed in the run of the first term for CMEE course.
The goal was to answer a general question: What mathematical models best fit an empirical data set?
We have gathered data from the TheMulQuaBio repository that looked at microbial growth rates. We have used Ordinary Least Squares and also Non Linear Least Squares to find models that fitted ther data and using Akaike's Information Criterion and Schwartz Criterion (BIC) to assist in the model selection procedure.
I have used a range of computational tools to build a totally reproducible workflow using R, Bash and Latex.
To run some of the scripts found in this folder, some specific packages are required. Please run the following code in R interpreter:
install.packages(c("tidyverse", "maps", "reshape2", "sqldf"))
If you have any questions regarding the materials please get in touch.
Vitor Ferreira
