Introduction

nf-core/deepmodeloptim is a bioinformatics end-to-end pipeline that performs statistically testing training procedures of machine learning models.

Deep learning model development in natural science is an empirical and costly process. Users must define a pre-processing pipeline, an architecture, find the best parameters for said architecture and iterate over this process.

Leveraging the power of Nextflow (polyglotism, container integration, scalable on the cloud), we propose STIMULUS, an open-source software built to automatize deep learning model development for genomics.

STIMULUS takes as input a user defined PyTorch model, a dataset, a configuration file to describe the pre-processing steps to be performed, and a range of parameters for the PyTorch model. It then transforms the data according to all possible pre-processing steps, finds the best architecture parameters for each of the transformed datasets, performs sanity checks on the models and train a minimal deep learning version for each dataset/architecture.

Those experiments are then compiled into an intuitive report, making it easier for scientists to pick the best design choice to be sent to large scale training.

Usage

Note

If you are new to Nextflow and nf-core, please refer to this page on how to set-up Nextflow. Make sure to test your setup with -profile test before running the workflow on actual data.

Now, you can run the pipeline using:

nextflow run nf-core/deepmodeloptim \
   -profile <docker/singularity/.../institute> \
   --input samplesheet.csv \
   --outdir <OUTDIR>

Warning

Please provide pipeline parameters via the CLI or Nextflow -params-file option. Custom config files including those provided by the -c Nextflow option can be used to provide any configuration except for parameters; see docs.

For more details and further functionality, please refer to the usage documentation and the parameter documentation.

Pipeline output

To see the results of an example test run with a full size dataset refer to the results tab on the nf-core website pipeline page. For more details about the output files and reports, please refer to the output documentation.

Code requirements

Data

The data is provided as a csv where the header columns are in the following format : name:type:class

name is user given (note that it has an impact on experiment definition).

type is either "input", "meta", or "label". "input" types are fed into the mode, "meta" types are registered but not transformed nor fed into the models and "label" is used as a training label.

class is a supported class of data for which encoding methods have been created, please raise an issue on github or contribute a PR if a class of your interest is not implemented

csv general example

input1:input:input_type	input2:input:input_type	meta1:meta:meta_type	label1:label:label_type	label2:label:label_type
sample1 input1	sample1 input2	sample1 meta1	sample1 label1	sample1 label2
sample2 input1	sample2 input2	sample2 meta1	sample2 label1	sample2 label2
sample3 input1	sample3 input2	sample3 meta1	sample3 label1	sample3 label2

csv specific example

mouse_dna:input:dna	mouse_rnaseq:label:float
ACTAGGCATGCTAGTCG	0.53
ACTGGGGCTAGTCGAA	0.23
GATGTTCTGATGCT	0.98

Model

In STIMULUS, users input a .py file containing a model written in pytorch (see examples in bin/tests/models)

Said models should obey to minor standards:

1. The model class you want to train should start with "Model", there should be exactly one class starting with "Model".

import torch
import torch.nn as nn

class SubClass(nn.Module):
    """
    a subclass, this will be invisible to Stimulus
    """

class ModelClass(nn.Module):
    """
    the PyTorch model to be trained by Stimulus, can use SubClass if needed
    """

class ModelAnotherClass(nn.Module):
    """
    uh oh, this will return an error as there are two classes starting with Model
    """

2. The model "forward" function should have input variables with the same names as the defined input names in the csv input file

import torch
import torch.nn as nn

class ModelClass(nn.Module):
    """
    the PyTorch model to be trained by Stimulus
    """
    def __init__():
        # your model definition here
        pass

    def forward(self, mouse_dna):
        output = model_layers(mouse_dna)

3. The model should include a batch named function that takes as input a dictionary of input "x", a dictionary of labels "y", a Callable loss function and a callable optimizer.

In order to allow batch to take as input a Callable loss, we define an extra compute_loss function that parses the correct output to the correct loss class.

import torch
import torch.nn as nn
from typing import Callable, Optional, Tuple

class ModelClass(nn.Module):
    """
    the PyTorch model to be trained by Stimulus
    """

    def __init__():
        # your model definition here
        pass

    def forward(self, mouse_dna):
        output = model_layers(mouse_dna)

    def compute_loss_mouse_rnaseq(self, output: torch.Tensor, mouse_rnaseq: torch.Tensor, loss_fn: Callable) -> torch.Tensor:
        """
        Compute the loss.
        `output` is the output tensor of the forward pass.
        `mouse_rnaseq` is the target tensor -> label column name.
        `loss_fn` is the loss function to be used.

        IMPORTANT : the input variable "mouse_rnaseq" has the same name as the label defined in the csv above.
        """
        return loss_fn(output, mouse_rnaseq)

    def batch(self, x: dict, y: dict, loss_fn: Callable, optimizer: Optional[Callable] = None) -> Tuple[torch.Tensor, dict]:
        """
        Perform one batch step.
        `x` is a dictionary with the input tensors.
        `y` is a dictionary with the target tensors.
        `loss_fn` is the loss function to be used.

        If `optimizer` is passed, it will perform the optimization step -> training step
        Otherwise, only return the forward pass output and loss -> evaluation step
        """
        output = self.forward(**x)
        loss = self.compute_loss_mouse_rnaseq(output, **y, loss_fn=loss_fn)
        if optimizer is not None:
            optimizer.zero_grad()
            loss.backward()
            optimizer.step()
        return loss, output

If you don't want to optimize the loss function, the code above can be written in a simplified manner

import torch
import torch.nn as nn
from typing import Callable, Optional, Tuple

class ModelClass(nn.Module):
    """
    the PyTorch model to be trained by Stimulus
    """

    def __init__():
        # your model definition here
        pass

    def forward(self, mouse_dna):
        output = model_layers(mouse_dna)

    def batch(self, x: dict, y: dict, optimizer: Optional[Callable] = None) -> Tuple[torch.Tensor, dict]:
        """
        Perform one batch step.
        `x` is a dictionary with the input tensors.
        `y` is a dictionary with the target tensors.
        `loss_fn` is the loss function to be used.

        If `optimizer` is passed, it will perform the optimization step -> training step
        Otherwise, only return the forward pass output and loss -> evaluation step
        """
        output = self.forward(**x)
        loss = nn.MSELoss(output, y['mouse_rnaseq'])
        if optimizer is not None:
            optimizer.zero_grad()
            loss.backward()
            optimizer.step()
        return loss, output

Model parameter search design

Experiment design

The file in which all information about how to handle the data before tuning is called an experiment_config. This file in .json format for now but it will be soon moved to .yaml. So this section could vary in the future.

The experiment_config is a mandatory input for the pipeline and can be passed with the flag --exp_conf followed by the PATH of the file you want to use. Two examples of experiment_config can be found in the examples directory.

Experiment config content description.

Credits

nf-core/deepmodeloptim was originally written by Mathys Grapotte.

We thank the following people for their extensive assistance in the development of this pipeline:

Contributions and Support

If you would like to contribute to this pipeline, please see the contributing guidelines.

For further information or help, don't hesitate to get in touch on the Slack #deepmodeloptim channel (you can join with this invite).

Citations

An extensive list of references for the tools used by the pipeline can be found in the CITATIONS.md file.

You can cite the nf-core publication as follows:

The nf-core framework for community-curated bioinformatics pipelines.

Philip Ewels, Alexander Peltzer, Sven Fillinger, Harshil Patel, Johannes Alneberg, Andreas Wilm, Maxime Ulysse Garcia, Paolo Di Tommaso & Sven Nahnsen.

Nat Biotechnol. 2020 Feb 13. doi: 10.1038/s41587-020-0439-x.