Team:USTC-Software/Proof
Module:Pathway finder
Our Path finder combines metabolic network databases, cell signaling pathway databases and gene interaction databases, to find a path between components which is usually genes or their associated proteins.[1] A click on the 'PATHFINDER' button will lead a way from one node to another through components registered in the databases that you want to make further investigation.
We took trp operon in E.coli for example, as the operon concept is considered one of the landmark events. The operon is regulated so that when tryptophan is present in the environment, the genes for tryptophan synthesis are not expressed.
Choose trpR and trpE genes as start node and end node. Path Finder gives 4 paths as Fig.1.
Fig.1 - A list of paths found by the 'Pathway finder' represented on the webpage.
The first route is considered as best-matching path, as it reveals mediate interactions between two genes. TrpR, the repressor for the trp operon is produced upstream by the trpR gene, which is constitutively expressed at a low level. When tryptophan is present, these tryptophan repressor dimers bind to tryptophan, allowing the repressor to bind to the operator, and leads to the transcription of trpLEDCBA. You can choose either one or all of these routes, which will be presented on the Pano. The latter circumstance is shown as Fig.2.
Fig.2 - A cute round network represented on the Pano.
Module:Simulation
Let us take an example of a biological reaction which involves three substances. We denote the concentration of each substance by yi and they satisfy the equations as Fig.3 shows. We input those control equations and initial values as well as the end time and the step length into the Simulation Pool (see as Fig.4).
Fig.3 - The control equations and the initial value of this example.
Fig.4 - The user interface of the 'Simulation' model.
The visualized output are given in Fig.5, while another output given by Mathematica is shown in Fig.6. Comparing these two graphs it is convinced that our ‘Simulation’ module gives out a quite acceptable answer. Although there are differences between these two pictures at the right side, these are caused by the chaos. The relatively right side of the solution is instable, which cannot be used as a reliable deduction.
Fig.5 - The result given by 'Simulation'.
Fig.6 - The result given by Mathematica.
Module:ABACUS
ABACUS is firstly developed by professor Liu Haiyan, to enlarge the functions of Biopano. The de novo design of amino acid sequence to fold into desired structures is a way to reach a more thorough understanding of how amino acid sequences encode protein structures and to supply methods for protein engineering. To overcome limitations in current computational models, they developed a comprehensive statistical energy function for protein design with a new general strategy and verify that it can complement and rival current well-established models. They established an experimental approach which can be used to efficiently assess or improve the foldability of designed proteins. To prove it, they report four de novo proteins for different targets, all experimentally verified to the well-folded, solved solution structures for two being excellent agreement with respective design targets.[2]
Unit Test
Our development is based on Git version control with multiple branches. All pushed commits will be automatically tested on our test server at once so that we can see the result soon. Besides unit tests, our test server will also process code analyzing to prevent fault codes and ensure that all classes and functions are well documented. When all is done, the new docker image will be generated automatically on DockerHub so that anyone can install and run our program with only one command.
Continuous integration and deployment ensure that we can fix bugs at the first time it appears. Unit tests ensure that our software performs all its intended function.
References
[1] Moisés Santillán, Michael C. Mackey. Dynamic regulation of the tryptophan operon: A modeling study and comparison with experimental data.
[2] Xiong P, Wang M, Zhou X, et al. Protein design with a comprehensive statistical energy function and boosted by experimental selection for foldability[J]. Nature communications, 2014, 5.
