Software Tools

Off-target Predictor

We exploit a software tool to filter experimental off-target sites from a huge number of candidates and ensure exclude almost all of the negative sequences. We have combined cas-OFFinder and our prediction model into an end-to-end pipeline. Users are just required to prepare the genome file and input their interested target sequence, the software will return the prediction result, high-confidence off-target sequences. Anyone can have access to our program at Github and use our software on the platform of Linux or macOS.

Maybe some users have many potential off-target sequences for distinct targets, they prefer inputting these sequence pairs directly and get the prediction on whether they are off-target sites or not. Therefore, we provide another prediction mode to meet such kind of demands.

Sample input and ouput for mode 1 (start from searching off-target candidates)

Sample input and ouput for mode 2 (directly input the sequence pairs)

Model collection

Maybe every iGEMer becomes anxious when he is looking for useful information from iGEM Team Wikis because it is hard to summarize critical points of project or model design from thousands of words. Therefore, the collection of previous models from iGEM Team Wikis, especially those focusing on hot spots in synthetic biology can develop a convenient method to learn from successful iGEM models for all iGEMers.

Since an awesome list is a list of awesome things curated by the community on Github, we intend to create an awesome list about outstanding iGEM models in some focused areas. As the first step, we consider to collect representative models concerning CRISPR system and machine learning which are mostly related to our work. We have collected several models about CRISPR system and machine learning in the last 3 year, and depicted Name, Type, Functions, Team, Link, Evaluation of the models. We can observe that most teams use experimental data and corresponding gene circuits to solve ordinary differential equations and build up the kinetic model for CRISPR system. Support Vector Machine, Random Forest, and Neural Network are often utilized in a prediction task (e.g. predict mutagenicity based on chemical substructures). Our present collection of models is already uploaded at Github.

In the future, we plan to take the jamboree as a great chance to gain insights into the current focus of iGEM teams' research. We will continue to organize iGEM models about key scientific issues on our Github page and consider to build a specialized database for all model materials from iGEM Team Wikis in next year. We hope to gradually increase the number of models collected and supplied online services such as a searching tool.

Source Code

All of our source code for prediction model, kinetics model, molecular dynamics model and graphic interpretation model are open source at Github . We are looking forward to receiving your STAR.

Acknowledgements

SJTU-BioX-Shanghai

contact us

Contact us: sjtuigem@gmail.com or sjtuigem@gmail.com