ICE tool is a CRISPR editing analysis tool that infers presence of indels and other mutation. ICE uses non-negative least squares regression to detect the presence or evidence of edits. In contrast to TIDE1, ICE can analyze insertions, deletions, HDR, multiplex edits, and base editing and is available under a open-source license for non-commercial use. ICE can also be used for analysis of other genome engineering methods, such as TALEN and homing endonucleases.
This document is intended for technical users who have prior experience with CRISPR editing analysis. For more detailed user documentation, please visit Synthego’s Help Center, where you can find our ICE User Guide and additional documentation.
A preprint Inference of CRISPR Edits from Sanger Trace Data2 provides an overview and empirical test of ICE on over 1,800 real world edits. We ask that you cite our paper if you use ICE in work that leads to publication.
We will continue to improve ICE, so please refer to the version number in your publication. The version number can be found by running ICE with the --version option.
- Control sample Sanger ab1 file
- Edited sample Sanger ab1 file
- Sequence of protospacer or gRNA target
Overall Editing efficiency, plots of distribution of edit types, plots of discordance (a calculation of signal agreement to control sequence), annotated Sanger traces of the region flanking the cut site, and JSON files containing the data for all of those plots.
Trace JSON The Sanger sequence traces for the region around the cut site for the edited and control samples are shown.
Discordance & indel distribution files
The Discord json shows the agreement of both the control and edited sample to the called base. A discordance of 1 indicates that there is zero signal in the base called at that particular position, whereas a discord of 0 would mean that all non-reference bases have zero signal (all signal agrees with called base). The alignment window is used to align the control sample to the edited sample, while the inference window denotes the subsection of data used for NNLS regression.
The indel distribution json shows the distribution of indel identified, summarized by length. Thus, two different -1 indels would be summarized to the same bin.
Sequence Contributions
Relative contribution of each sequence (normalized) -------------------------------------------------------- 0.3006 -1[g1] CCCAACACAACCAGTTGCAGGCGCC|-CATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.1996 0[g1] CCCAACACAACCAGTTGCAGGCGCC|CCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.1818 1[g1] CCCAACACAACCAGTTGCAGGCGCC|nCCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.1128 -2[g1] CCCAACACAACCAGTTGCAGGCGC-|-CATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0541 2[g1] CCCAACACAACCAGTTGCAGGCGCC|nnCCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0317 -1[g1] CCCAACACAACCAGTTGCAGGCGC-|CCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0296 -4[g1] CCCAACACAACCAGTTGCAGGC---|-CATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0185 -3[g1] CCCAACACAACCAGTTGCAGGC---|CCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0134 -19[g1] CCCAACACAACCAGT----------|---------GCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0078 -16[g1] CCCAACACAACCAGTTG--------|--------AGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0067 -18[g1] CCCAACACAA---------------|---TGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0060 -20[g1] CCCAACACAA---------------|-----GTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0053 -16[g1] CCCAACACAACCAGTTGCAGGC---|-------------CAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0042 -16[g1] CCCAACACAA---------------|-CATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0032 -15[g1] CCCAACACAACC-------------|--ATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0028 -4[g1] CCCAACACAACCAGTTGCAGGCGCC|----GGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0021 -20[g1] CCCAACACAACCA------------|--------AGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0012 -17[g1] CCCAACACAA---------------|--ATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA 0.0005 -13[g1] CCCAACACAACC-------------|CCATGGTGAGCATCAGCCTCTGGGTGGCCCTCCCTCTGGGCCTCGGGTATTTATGGAGCTGGATCCAAGGTCACATGCTTGTTCATGAGCTCTCAGGCA
The first column indicates the proportion of that sequence inferred in the pool. The second column is a summarized identity indicating the size of the indel and which guide. The third column is a human-readable representation of the sequence with dashes indicating deletions and 'n' indicating insertions.
Additional files such as alignment verification are generated for each sample.
A hosted free version of the ICE tool is available online at https://ice.synthego.com. The online ICE tool supports batch analysis, figure generation, and error handling & sample QC.
The source code behind the core ICE analysis is open source and free to use for non-commercial applications. Commercial use and other licensing options are available. For details, see LICENSE.
Synthego ICE can be installed as a docker container or directly via pip. Additional developer instructions are located in DEVELOP.md. All examples below use test data found in ice/tests/test_data. The test file (./ice/tests/test_data/batch_example.xlsx) is an example of how to specify batch inputs.
Install into your favorite python3 virtual environment (virtualenv, conda).
conda create --name ice_env python=3 # create a python3 virtual environment
source activate ice_env # activate the virtual environment
pip install sythego_ice # install synthego ice from pipAfter installation, you can use Synthego ICE as a module (see python_example.py) or directly via command line.
synthego_ice
usage: synthego_ice [-h] --control CONTROL --edited EDITED --target TARGET
[--out OUT] [--donor DONOR] [--verbose] [--version]
Analyze Sanger reads to Infer Crispr Edit outcomes
optional arguments:
-h, --help show this help message and exit
--control CONTROL The wildtype / unedited ab1 file (REQUIRED)
--edited EDITED The edited ab1 file (REQUIRED)
--target TARGET Target sequence(s) (17-23 bases, RNA or DNA, comma
separated), (REQUIRED)
--out OUT Output base path (Defaults to ./results/single)
--donor DONOR Donor DNA sequence for HDR (Optional)
--verbose
--version show program's version number and exit
synthego_ice_batch
usage: synthego_ice_batch [-h] --in INPUT [--out OUT] --data DATA [--verbose]
[--line LINE] [--allprops] [--version]
Analyze Sanger reads to infer crispr edit outcomes
optional arguments:
-h, --help show this help message and exit
--in INPUT Input definition file in Excel xlsx format (required)
--out OUT Output directory path (defaults to .)
--data DATA Data path, where .ab1 files are located (required)
--verbose Display verbose output
--line LINE Only run specified line in the Excel xlsx definition file
--allprops Output all Edit Proposals, even if they have zero contribution
--version show program's version number and exit
After installing via pip, grab the example data by cloning this repository:
git clone git@github.com:synthego-open/ice.git ice
cd ice # change into the ice directoryAnalyzing a single sample
synthego_ice \
--control ./ice/tests/test_data/good_example_control.ab1 \
--edited ./ice/tests/test_data/good_example_edited.ab1 \
--target AACCAGTTGCAGGCGCCCCA \
--out results/testing \
--verboseWhen complete, you'll find the ICE analysis outputs in the ./results folder.
Analyzing multiple samples (batch analysis)
synthego_ice_batch \
--in ./ice/tests/test_data/batch_example.xlsx \
--out ./results/ \
--data ./ice/tests/test_data/
--verboseWhen complete, you'll find the ICE analysis outputs in the ./results folder.
- Docker http://docker.com
From a command line, grab the latest version of Synthego ICE from Docker Hub.
docker pull synthego/iceAfter installation, you'll be able to run Synthego ICE from the docker container.
Grab the example data by cloning this repository:
git clone git@github.com:synthego-open/ice.git ice
cd ice # change into the ice directoryAnalyzing a single sample
docker run -it -v ${PWD}:/data -w /ice -i ice:latest \
python ice_analysis_single.py \
--control /data/ice/tests/test_data/good_example_control.ab1 \
--edited /data/ice/tests/test_data/good_example_edited.ab1 \
--target AACCAGTTGCAGGCGCCCCA \
--out /data/results/testing \
--verboseWhen complete, you'll find the ICE analysis outputs in the ./results folder.
Analyzing multiple samples (batch analysis)
docker run -it -v ${PWD}:/data -w /ice -i ice:latest \
python ice_analysis_batch.py \
--in /data/ice/tests/test_data/batch_example.xlsx \
--data /data/ice/tests/test_data/ \
--out /data/results/ \
--verboseWhen complete, you'll find the ICE analysis outputs in the ./results folder.
Pull requests are welcome. Please follow the below steps to ensure your work is merged as efficiently as possible.
- Make a Github issue outlining the bug you aim to fix or the feature you want to add. This prevents redundant work and lets us reach an agreement on your proposal before you put significant effort into it.
- Fork the repository and create your branch from master.
- Code your feature or bug fix.
- Add tests for you new code and make sure all other tests still pass.
- Submit a pull request referencing the issue from step 1.
Tests are written using pytest. Test files can be found in ice/tests.
- Run all tests (can take a few minutes):
$ pytest
- Run a specific test:
$ pytest ice/tests/{{filename}}.py::{{test_function_name}}$ pytest ice/tests/test_utility.py::test_sequence_util
- Run all tests with coverage report:
$ py.test --cov-report term-missing --cov=ice ice/tests
[1] Eva K. Brinkman, Tao Chen, Mario Amendola, and Bas van Steensel. Easy quantitative assessment of genome editing by sequence trace decomposition.Nucleic Acids Res. 2014 Dec 16; 42(22): e168. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4267669/
[2] Hsiau et. al, Inference of CRISPR Edits from Sanger Trace Data. BioArxiv. 2018 https://www.biorxiv.org/content/early/2018/01/20/251082
Copyright 2018 Synthego Corporation All Rights Reserved
The Synthego ICE software was developed at Synthego Corporation.
Permission to use, copy, modify and distribute any part of Synthego ICE for educational, research and non-profit purposes, without fee, and without a written agreement is hereby granted, provided that the above copyright notice, this paragraph and the following paragraphs appear in all copies.
Those desiring to incorporate this Synthego ICE software into commercial products or use for commercial purposes should contact Synthego support at Ph: (888) 611-6883 ext:1, E-MAIL: support@synthego.com.
In no event shall Synthego Corporation be liable to any party for direct, indirect, special, incidental, or consequential damages, including lost profits, arising out of the use of Synthego ICE, even if Synthego Corporation has been advised of the possibility of such damage.
The Synthego ICE tool provided herein is on an "as is" basis, and the Synthego Corporation has no obligation to provide maintenance, support, updates, enhancements, or modifications. The Synthego Corporation makes no representations and extends no warranties of any kind, either implied or express, including, but not limited to, the implied warranties of merchantability or fitness for a particular purpose, or that the use of Synthego ICE will not infringe any patent, trademark or other rights.