MPRA-DragoNN/DeepFactorizedModel

Authors: Rajiv Movva, Surag Nair

License: MIT

Contributed by: Rajiv Movva, Surag Nair

Type: None

Trained on: Sharpr-MPRA dataset. chr8 validation, chr18 test. other chromosomes train.

Deep factorized convolutional neural network for predicting Sharpr-MPRA activity of arbitrary 145bp sequences. Architecture based on https://doi.org/10.1101/229385.

Create a new conda environment with all dependencies installed

kipoi env create MPRA-DragoNN/DeepFactorizedModel
source activate kipoi-MPRA-DragoNN__DeepFactorizedModel

Test the model

kipoi test MPRA-DragoNN/DeepFactorizedModel --source=kipoi

Make a prediction

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

Create a new conda environment with all dependencies installed

kipoi env create MPRA-DragoNN/DeepFactorizedModel
source activate kipoi-MPRA-DragoNN__DeepFactorizedModel

Get the model

import kipoi
model = kipoi.get_model('MPRA-DragoNN/DeepFactorizedModel')

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('MPRA-DragoNN/DeepFactorizedModel')

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:mpra-dragonn-slim

Get the full sized docker image

docker pull kipoi/kipoi-docker:mpra-dragonn

Get the activated conda environment inside the container

docker run -it kipoi/kipoi-docker:mpra-dragonn-slim

Test the model

docker run kipoi/kipoi-docker:mpra-dragonn-slim kipoi test MPRA-DragoNN/DeepFactorizedModel --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:mpra-dragonn-slim \
kipoi get-example MPRA-DragoNN/DeepFactorizedModel -o /app/example 
docker run -v $PWD/kipoi-example:/app/ kipoi/kipoi-docker:mpra-dragonn-slim \
kipoi predict MPRA-DragoNN/DeepFactorizedModel \
--dataloader_args='{'intervals_file': '/app/example/intervals_file', 'fasta_file': '/app/example/fasta_file'}' \
-o '/app/MPRA-DragoNN_DeepFactorizedModel.example_pred.tsv' 
# check the results
head $PWD/kipoi-example/MPRA-DragoNN_DeepFactorizedModel.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 MPRA-DragoNN/DeepFactorizedModel -o example
kipoi predict MPRA-DragoNN/DeepFactorizedModel \
--dataloader_args='{"intervals_file": "example/intervals_file", "fasta_file": "example/fasta_file"}' \
-o 'MPRA-DragoNN_DeepFactorizedModel.example_pred.tsv' \
--singularity 
# check the results
head MPRA-DragoNN_DeepFactorizedModel.example_pred.tsv

Schema

Inputs

Single numpy array

Name: None

Shape: (145, 4)

Doc: 145bp one-hot encoded ACGT sequences (e.g. [1,0,0,0] = 'A')

Targets

Single numpy array

Name: None

Shape: (12,)

Doc: predicts 12 tasks: k562 minP replicate 1, k562 minP replicate 2, k562 minP pooled, k562 sv40p replicate 1, k562 sv40p replicate 2, k562 sv40p pooled, hepg2 minP replicate 1, hepg2 minP replicate 2, hepg2 minP pooled, hepg2 sv40p replicate 1, hepg2 sv40p replicate 2, hepg2 sv40p pooled.

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

Model dependencies

conda:

cython=0.28.5
python=3.7
h5py=2.8.0
pip=20.3.3
keras=2.3
tensorflow=1.14

pip:

protobuf==3.20

Dataloader dependencies

conda:

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

pip:

kipoiseq