Taiyaki is research software for training models for basecalling Oxford Nanopore reads.
Oxford Nanopore's devices measure the flow of ions through a nanopore, and detect changes in that flow as molecules pass through the pore. These signals can be highly complex and exhibit long-range dependencies, much like spoken or written language. Taiyaki can be used to train neural networks to understand the complex signal from a nanopore device, using techniques inspired by state-of-the-art language processing.
Taiyaki is used to train the models used to basecall DNA and RNA found in Oxford Nanopore's Guppy basecaller (version 2.2 at time of writing). This includes the flip-flop models, which are trained using a technique inspired by Connectionist Temporal Classification (Graves et al 2006).
- Prepare data for training basecallers by remapping signal to reference sequence
- Train neural networks for flip-flop basecalling and squiggle prediction
- Export basecaller models for use in Guppy
Taiyaki is built on top of pytorch and is compatible with Python 3.5 or later. It is aimed at advanced users, and it is an actively evolving research project, so expect to get your hands dirty.
- Install system prerequisites
- Walk through
- Guppy compatibility
- Environment variables
- Running on UGE
Install system prerequisites
To install required system packages on ubuntu 16.04:
sudo make deps
Other linux platforms may be compatible, but are untested.
In order to accelerate model training with a GPU you will need to install CUDA (which should install nvcc and add it to your path.) See instructions from NVIDIA and the CUDA section below.
Taiyaki also makes use of the OpenMP extensions for multi-processing. These are supported by the system installed compiler on most modern Linux systems but require a more modern version of the clang/llvm compiler than that installed on MacOS machines. Support for OpenMP was adding in clang/llvm in version 3.7 (see http://llvm.org or use brew). Alternatively you can install GCC on MacOS using homebrew.
Some analysis scripts require a recent version of the BWA aligner.
Windows is not supported.
CUDA before proceeding.NOTE If you intend to use Taiyaki with a GPU, make sure you have installed and set up
Install Taiyaki in a new virtual environment
We recommend installing Taiyaki in a self-contained virtual environment.
The following command creates a complete environment for developing and testing Taiyaki, in the directory venv:
Taiyaki will be installed in development mode so that you can easily test your changes.
You will need to run
source venv/bin/activate at the start of each session when you want to use this virtual environment.
Install Taiyaki system-wide or into activated Python environment
Taiyaki can be installed from source using either:
python3 setup.py install python3 setup.py develop #[development mode](http://setuptools.readthedocs.io/en/latest/setuptools.html#development-mode)
Alternatively, you can use pip with either:
pip install path/to/taiyaki/repo pip install -e path/to/taiyaki/repo #[development mode](http://setuptools.readthedocs.io/en/latest/setuptools.html#development-mode)
Tests can be run as follows:
make workflow #runs scripts which carry out the workflow for basecall-network training and for squiggle-predictor training make acctest #runs acceptance tests make unittest #runs unit tests
If Taiyaki has install in a virtual environment, it will have to activated before running tests:
source venv/bin/activate. To deactivate, run
Walk throughs and further documentation
For a walk-through of Taiyaki model training, including how to obtain sample training data, see docs/walkthrough.rst.
For an example of training a modifed base model, see docs/modbase.rst.
Using the workflow Makefile
The file at workflow/Makefile can be used to direct the process of generating ingredients for training and then running the training itself.
For example, if we have a directory read_dir containing fast5 files, and a fasta file refs.fa containing a ground-truth reference sequence for each read, we can (from the Taiyaki root directory) use the command line
make -f workflow/Makefile MAXREADS=1000 \ READDIR=read_dir USER_PER_READ_REFERENCE_FILE=refs.fa \ DEVICE=3 train_remapuser_ref
This will place the training ingredients in a directory RESULTS/training_ingredients and the training output (including logs and trained models) in RESULTS/remap_training, using GPU 3 and only reading the first 1000 reads in the directory. The fast5 files may be single or multi-read.
Using command line options to make, it is possible to change various other options, including the directory where the results go. Read the Makefile to find out about these options. The Makefile can also be used to follow a squiggle-mapping workflow.
The paragraph below describes the steps in the workflow in more detail.
Steps from fast5 files to basecalling
The script bin/prepare_mapped_reads.py prepares a file containing mapped signals. This mapped signal file is then used to train a basecalling model.
The simplest workflow looks like this. The flow runs from top to bottom and lines show the inputs required for each stage. The scripts in the Taiyaki package are shown, as are the files they work with.
fast5 files / \ / \ / \ / generate_per_read_params.py | | | | fasta with reference | per-read-params file sequence for each read | (tsv, contains shift, (produced with get_refs_from_sam.py | scale, trim for each read) or some other method) \ | / \ | / \ | / \ | / \ | / \ | / prepare_mapped_reads.py (also uses remapping flip-flop model from models/) | | mapped-signal-file (hdf5) | | train_flipflop.py (also uses definition of model to be trained) | | trained flip-flop model | | dump_json.py | | json model definition (suitable for use by Guppy)
Each script in bin/ has lots of options, which you can find out about by reading the scripts. Basic usage is as follows:
bin/generate_per_read_params.py <directory containing fast5 files> --output <name of output per_read_tsv file> bin/get_refs_from_sam.py <genomic references fasta> <one or more SAM/BAM files> --output <name of output reference_fasta> bin/prepare_mapped_reads.py <directory containing fast5 files> <per_read_tsv> <output mapped_signal_file> <file containing model for remapping> <reference_fasta> bin/train_flipflop.py --device <digit specifying GPU> --chunk_logging_threshold 0 <pytorch model definition> <output directory for checkpoints> <mapped-signal files to train with>
We suggest using the chunk_logging_threshold 0 to begin with. This results in all chunks (including rejected chunks) being logged in a tsv file in the training directory. This chunk log can be useful for diagnosing problems, but can get quite large, so may be turned off for very long training runs.
Some scripts mentioned also have a useful option --limit which limits the number of reads to be used. This allows a quick test of a workflow.
Preparing a training set
prepare_mapped_reads.py script prepares a data set to use to train a new basecaller. Each member of this data set contains:
- The raw signal for a complete nanopore read (lifted from a fast5 file)
- A reference sequence that is the "ground truth" for the that read
- An alignment between the signal and the reference
As input to this script, we need a directory containing fast5 files (either single-read or multi-read) and a fasta file that contains the ground-truth reference for each read. In order to match the raw signal to the correct ground-truth sequence, the IDs in the fasta file should be the unique read ID assigned by MinKnow (these are the same IDs that Guppy uses in its fastq output). For example, a record in the fasta file might look like:
The recommended way to produce this fasta file is as follows:
- Align Guppy fastq basecalls to a reference genome using Guppy Aligner or Minimap. This will produce one or more SAM files.
- Use the
get_refs_from_sam.pyscript to extract a snippet of the reference for each mapped read. You can filter reads by coverage.
The final input required by
prepare_mapped_signal.py is a pre-trained basecaller model, which is used to determine the alignment between raw signal and reference sequence.
An example of such a model (for DNA sequenced with pore r9) is provided at
This does make the entire training process somewhat circular: you need a model to train a model.
However, the new training set can be somewhat different from the data that the remapping model was trained on and things still work out.
So, for example, if your samples are a bit weird and whacky, you may be able to improve basecall accuracy by retraining a model with Taiyaki.
Internally, we use Taiyaki to train basecallers after incremental pore updates, and as a research tool into better basecalling methods.
Taiyaki is not intended to enable training basecallers from scratch for novel nanopores.
If it seems like remapping will not work for your data set, then you can use alternative methods
so long as they produce data conformant with this format.
Taiyaki comes with a script to perform flip-flop basecalling using a GPU. This script requries CUDA and cupy to be installed.
bin/basecall.py <directory containing fast5s> <model checkpoint> > <output fasta>
A limited range of models can also be used with Guppy, which will provide better performance and stability. See the section on Guppy compatibility for more details.
Taiyaki enables the training of models to predict the presence of modified bases (a.k.a. non-canonical or alternative bases) alongside the standard flip-flop canonical base probabilities via an alteration to the model architecture (model architecture referred to as categorical modifications, or
cat_mod for short).
This alteration results in a second stream of data from the neural network which represents the probability that any base is canonical or modified (potentially include any number of modifications).
A number of adjustments to the training workflow are required to train a modified base model.
These adjustments begin with the “FASTA with reference sequence for each read” which is input to the
This FASTA file should contain ground truth per-read references annotated with modified base locations.
The single letter codes used to represent modified bases can be arbitrary and are defined using the
--mod command line argument to the
--mod argument takes three parameters: the letter representing the modified base, the letter representing its canonical representation, and a long name.
--mod argument should be repeated once for each modification.
For example, to encode 5-methyl-cytosine and 6-methyl-adenosine with the single letter codes
Y respectively, the following commandline arguments would be added
--mod Z C 5mC --mod Y A 6mA.
These values will be stored in the prepared signal mapped HDF5 output file for use in training downstream.
mapped-signal-file is passed into the
train_mod_flipflop.py command (as opposed to
train_flipflop.py from standard workflow).
This script requires a
cat_mod model to be provided (e.g.
This script also provides a number of arguments specific to training a
--mod_factor argument controls the proportion of the training loss attributed to the modified base output stream in comparison to the canonical base output stream.
When training a model from scratch it is generally recommended to set this factor to a lower value (
0.01 for example) to train the model to call canonical bases and then restart training with the default,
1, value in order to train the model to identify modified bases.
Modified base models can be used in megalodon (release imminent) to call modified bases anchored to a reference.
'Ab initio' is an alternative entry point for Taiyaki that obtains acceptable models with fewer input requirements, particularly it does not require a previously trained model.
The input for ab initio training is a set of signal-sequence pairs:
- Fixed length chunks from reads
- A reference sequence trimmed for each chunk.
The models produced are not as accurate as normal training process but can be used to bootstrap it.
The process is described in the abinitio walk-through.
In order to train a model that is compatible with Guppy (version 2.2 at time of writing), we recommend that you
use the model defined in
models/mGru_flipflop.py and that you call
train_flipflop.py --size 256 --stride 2 --winlen 19 mGru_flipflop.py <other options...>
You should then be able to export your checkpoint to json (using bin/dump_json.py) that can be used to basecall with Guppy.
See Guppy documentation for more information on how to do this.
Key options include selecting the Guppy config file to be appropriate for your application, and passing the complete path of your .json file.
guppy_basecaller --input_path /path/to/input_reads --save_path /path/to/save_dir --config dna_r9.4.1_450bps_flipflop.cfg --model path/to/model.json --device cuda:1
Certain other model architectures may also be Guppy-compatible, but it is hard to give an exhaustive list and so we recommend you contact us to get confirmation.
We are working on adding basecalling functionality to Taiyaki itself to support a wider range of models.
The environment variables
OPENBLAS_NUM_THREADS can have an impact on performance.
The optimal value will depend on your system and on the jobs you are running, so experiment.
As a starting point, we recommend:
In order to use a GPU to accelerate model training, you will need to ensure that CUDA is installed (specifically nvcc) and that CUDA-related environment variables are set.
This should be done before running
make install described above. If you forgot to do this, just run
make install again once everything is set up.
The Makefile will try to detect which version of CUDA is present on your system, and install matching versions of pytorch and cupy.
To see what version of CUDA will be detected and which torch and cupy packages will be installed you can run:
Expert users can override the detected versions on the command line. For example, you might want to do this if you are building Taiyaki on one machine to run on another.
# Force CUDA version 8.0 CUDA=8.0 make install # Override torch package, and don't install cupy at all TORCH=my-special-torch-package CUPY= make install
Users who install Taiyaki system-wide or into an existing activated Python environment will need to make sure CUDA and a corresponding version of PyTorch have been installed.
During training, if this error occurs:
AttributeError: module 'torch._C' has no attribute '_cuda_setDevice'
or any other error related to the device, it suggests that you are trying to use pytorch's CUDA functionality but that CUDA (specifically nvcc) is either not installed or not correctly set up.
-bash: nvcc: command not found
nvcc is not installed or it is not on your path.
Ensure that you have installed CUDA (check NVIDIA's intructions) and that the CUDA compiler
nvcc is on your path.
To place cuda on your path enter the following:
export PATH=$PATH:/usr/local/cuda/bin export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64
Once CUDA is correctly configured and you are installing Taiyaki in a new virtual environment (as recommended), you may need to run
make install again to ensure that you have the correct pytorch package to match your CUDA version.
Running on a UGE cluster
There are two things to get right: (a) installing with the correct CUDA version, and (b) executing with the correct choice of GPU.
(a) It is important that when the package is installed, it knows which version of the CUDA compiler is available on the machine where it will be executed. When running on a UGE cluster we might want to do installation on a different machine from execution. There are two ways of getting around this. You can qlogin to a node which has the same resources as the execution node, and then install using that machine:
qlogin -l h=<nodename> cd <taiyaki_directory> make install
...or you can tell Taiyaki at the installation stage which version of CUDA to use. For example
CUDA=8.0 make install
(b) When executing on a UGE cluster you need to make sure you run on a node which has GPUs available, and then tell Taiyaki to use the correct GPU.
You tell the system to wait for a node which has an available GPU by adding the option -l gpu=1 to your qsub command. To find out which GPU has been allocated to your job, you need to look at the environment variable SGE_HGR_gpu. If it has the value cuda0, then use GPU number 0, and if it has the value cuda1, then use GPU 1. The command line option --device (used by train_flipflop.py accepts inputs such as 'cuda0' or 'cuda1' or integers 0 or 1, so SGE_HGR_gpu can be passed straight into the --device option.
The easy way to achieve this is with a Makefile like the one in the directory workflow. This Makefile contains comments which will help users run the package on a UGE system.
The misc directory contains several scripts that are useful for working out where things went wrong (or understanding why they went right).
Graphs showing the information in mapped read files can be plotted using the script plot_mapped_signals.py A graph showing the progress of training can be plotted using the script plot_training.py
When train_flipflop.py is run with the option --chunk_logging_threshold 0 then all chunks examined are logged (including those used to set chunk filtering parameters and those rejected for training). The script plot_chunklog.py plots several pictures that make use of this logged information.
This is a research release provided under the terms of the Oxford Nanopore Technologies' Public Licence. Research releases are provided as technology demonstrators to provide early access to features or stimulate Community development of tools. Support for this software will be minimal and is only provided directly by the developers. Feature requests, improvements, and discussions are welcome and can be implemented by forking and pull requests. However much as we would like to rectify every issue and piece of feedback users may have, the developers may have limited resource for support of this software. Research releases may be unstable and subject to rapid iteration by Oxford Nanopore Technologies.
© 2019 Oxford Nanopore Technologies Ltd. Taiyaki is distributed under the terms of the Oxford Nanopore Technologies' Public Licence.