Computational tools to improve genome engineering applications


The GT-Scan suite contains computational tools for efficiently solving the optimisation problem of identifying the most suitable target for genome engineering approaches, which requires to maximise on-target activity (guide efficiency) while also to minimise potential off-target effects (guide specificity).

The GT-Scan suite is a machine learning framework that is continuously being improved by CSIRO’s bioinformatics researchers. As such, CSIRO’s own novel research findings are incorporated allowing it to be at the cutting edge:

  • Incorporating chromatin marks to improve activity prediction accuracy by 30%
  • Using genomic variants, e.g. SNPs, to identify target sites for the sample at hand
  • Predicting on- and off-target activity to model experimentally observed efficiency
  • Computationally powerful enough for genome-wide application


The GT-Scan tools are web-services allowing researchers to get results for their experiments directly through the browser. However, all tools also have an API, which allows automatic, reproducible and interoperable genomic research, e.g. by invoking the API through a Jupyter or AWS SageMaker notebook.


The GT-Scan suite contains:

  • GT-Scan: efficient off-target search tool supporting more than 50 species including plant, fungi, fish and amphibians as well as a wide range of genome editing approaches (CRISPR, TALANs, Zincfinger) – available
  • TUSCAN: On-target prediction method based on random forest and trained on CRISPR/Cas9 sequencing data – available
  • CHROMscan: On-target prediction tool like TUSCAN but incorporating chromatin information
  • GT-Scan2: joining GT-Scan and CHROMscan – available
  • VARSCOT: SNP-aware off-target finder and off-target activity scorer for CRISPR/Cas9 – in preparation
  • Cas12a: on-target activity scorer for CRISPR/Cas12a (formerly known as Cpf1) – in preparation
  • HDR: on-target activity scorer for CRISPR-based Knock-in approaches – in preparation

Publications and Media