DeepSEA/beluga

Authors: Jian Zhou , Olga G. Troyanskaya

License: Non-comercial

Contributed by: Jian Zhou , Olga G. Troyanskaya

Cite as: https://doi.org/10.1038/s41588-018-0160-6

Type: None

Trained on: Chromosome 8 and 9 were excluded from training, and the rest of the autosomes were used for training and validation. 4,000 samples on chromosome 7 spanning the genomic coordinates 30,508,751-35,296,850 were used as the validation set.

Source files

This model (DeepSEA Beluga) is a part of the ExPecto model for predicting expression from sequence. The model itself is a deep convolutional network model of TF binding, DNase accessibility and histone marks. Comparing to DeepSEA, this model contains 2x number of convolution layers, takes 2000bp input, and expanded the histone mark collections to all of Roadmap Epigenomics release.

Create a new conda environment with all dependencies installed

kipoi env create DeepSEA/beluga
source activate kipoi-DeepSEA__beluga

Test the model

kipoi test DeepSEA/beluga --source=kipoi

Make a prediction

kipoi get-example DeepSEA/beluga -o example
kipoi predict DeepSEA/beluga \
  --dataloader_args='{"intervals_file": "example/intervals_file", "fasta_file": "example/fasta_file"}' \
  -o '/tmp/DeepSEA|beluga.example_pred.tsv'
# check the results
head '/tmp/DeepSEA|beluga.example_pred.tsv'

Create a new conda environment with all dependencies installed

kipoi env create DeepSEA/beluga
source activate kipoi-DeepSEA__beluga

Get the model

import kipoi
model = kipoi.get_model('DeepSEA/beluga')

Make a prediction for example files

pred = model.pipeline.predict_example(batch_size=4)

Use dataloader and model separately

# Download example dataloader kwargs
dl_kwargs = model.default_dataloader.download_example('example')
# Get the dataloader and instantiate it
dl = model.default_dataloader(**dl_kwargs)
# get a batch iterator
batch_iterator = dl.batch_iter(batch_size=4)
for batch in batch_iterator:
    # predict for a batch
    batch_pred = model.predict_on_batch(batch['inputs'])

Make predictions for custom files directly

pred = model.pipeline.predict(dl_kwargs, batch_size=4)

Get the model

library(reticulate)
kipoi <- import('kipoi')
model <- kipoi$get_model('DeepSEA/beluga')

Make a prediction for example files

predictions <- model$pipeline$predict_example()

Use dataloader and model separately

# Download example dataloader kwargs
dl_kwargs <- model$default_dataloader$download_example('example')
# Get the dataloader
dl <- model$default_dataloader(dl_kwargs)
# get a batch iterator
it <- dl$batch_iter(batch_size=4)
# predict for a batch
batch <- iter_next(it)
model$predict_on_batch(batch$inputs)

Make predictions for custom files directly

pred <- model$pipeline$predict(dl_kwargs, batch_size=4)

Get the docker image

docker pull kipoi/kipoi-docker:sharedpy3keras2tf2-slim

Get the full sized docker image

docker pull kipoi/kipoi-docker:sharedpy3keras2tf2

Get the activated conda environment inside the container

docker run -it kipoi/kipoi-docker:sharedpy3keras2tf2-slim

Test the model

docker run kipoi/kipoi-docker:sharedpy3keras2tf2-slim kipoi test DeepSEA/beluga --source=kipoi

Make prediction for custom files directly

# Create an example directory containing the data
mkdir -p $PWD/kipoi-example 
# You can replace $PWD/kipoi-example with a different absolute path containing the data 
docker run -v $PWD/kipoi-example:/app/ kipoi/kipoi-docker:sharedpy3keras2tf2-slim \
kipoi get-example DeepSEA/beluga -o /app/example 
docker run -v $PWD/kipoi-example:/app/ kipoi/kipoi-docker:sharedpy3keras2tf2-slim \
kipoi predict DeepSEA/beluga \
--dataloader_args='{'intervals_file': '/app/example/intervals_file', 'fasta_file': '/app/example/fasta_file'}' \
-o '/app/DeepSEA_beluga.example_pred.tsv' 
# check the results
head $PWD/kipoi-example/DeepSEA_beluga.example_pred.tsv

Install apptainer

https://apptainer.org/docs/user/main/quick_start.html#quick-installation-steps

Make prediction for custom files directly

kipoi get-example DeepSEA/beluga -o example
kipoi predict DeepSEA/beluga \
--dataloader_args='{"intervals_file": "example/intervals_file", "fasta_file": "example/fasta_file"}' \
-o 'DeepSEA_beluga.example_pred.tsv' \
--singularity 
# check the results
head DeepSEA_beluga.example_pred.tsv

Schema

Inputs

Single numpy array

Name: seq

Shape: (4, 1, 2000)

Doc: DNA sequence

Targets

Single numpy array

Name: TFBS_DHS_probs

Shape: (2002,)

Doc: Probability of a specific epigentic feature

Dataloader

Defined as: kipoiseq.dataloaders.SeqIntervalDl

Doc: Dataloader for a combination of fasta and tab-delimited input files such as bed files. The dataloader extracts regions from the fasta file as defined in the tab-delimited `intervals_file` and converts them into one-hot encoded format. Returned sequences are of the type np.array with the shape inferred from the arguments: `alphabet_axis` and `dummy_axis`.

Authors: Ziga Avsec , Roman Kreuzhuber

Type: Dataset

License: MIT

Arguments

intervals_file : bed3+<columns> file path containing intervals + (optionally) labels

fasta_file : Reference genome FASTA file path.

num_chr_fasta (optional): True, the the dataloader will make sure that the chromosomes don't start with chr.

label_dtype (optional): None, datatype of the task labels taken from the intervals_file. Example: str, int, float, np.float32

use_strand (optional): reverse-complement fasta sequence if bed file defines negative strand. Requires a bed6 file

ignore_targets (optional): if True, don't return any target variables

Model dependencies

conda:

python=3.8
pip=22.0.4
h5py=3.9.0
pytorch::pytorch=2.0.1
cython=3.0.0

pip:

kipoi
kipoiseq

Dataloader dependencies

conda:

bioconda::pybedtools
bioconda::pyfaidx
bioconda::pyranges
numpy
pandas

pip:

kipoiseq