Team:UiOslo Norway/Contribution


Disclaimer

Due to restrictions imposed by the COVID-19 situation we were unable to perform any BioBrick contribution or lab work of any kind. The contribution we were able to do is of a computational nature.

Extendable code on GitHub

Our computational work consist of three pieces of software; modeling, analysis and visualization. Currently these pieces of code allows the user to run simulations of fish schooling with different parameters. They can then analyse these simulations with a neural net who tries to classify them into two different classes. We have been investigating “sick” and “healthy”, lastly our visualization software allows the user to see the simulation in 3 dimensions. We have done our best to document and organize the code properly, making it easily extendable.

Our software can be used by novice users, to change parameters for the model one can simply edit a .txt file and hit run. Everything should be ready to go as long as the user has compilers and writes a path to the required CSV in blender.

In the extendibility department our software pieces complement each other. If the user wants to introduce new behaviours, he can easily verify the performance by visualizing the simulation. If he wants to investigate new measurements, then their performance can be evaluated with our analysis code.

Blender and extension of our visualization work to other projects

Blender is an open-source animation and 3D modeling software. It supports scripting via a Python API (Blender Python, bpy) which allows for automation of large tasks and direct interaction of Blender functions. We believe that the blenderScript.py in (igem gitbhub) can be used as a base to build a pipeline to import large datasets and visualize them in Blender.

Seeing as Blender has built in support for Protein Data Bank files (.pdb), we believe this script can be modified to automate import of a large number of proteins in a scene with a single click.