DeepLiver/DeepLiver_Accessibility

Authors: Carmen Bravo , Stein Aerts

License: Other / Non-commercial (see LICENSE.txt)

Contributed by: Carmen Bravo , Stein Aerts

Cite as: Bravo
González-Blas
Carmen.
(2022).
Enhancer
grammar
of
liver
cell
types
and
hepatocyte
zonation
states.
https://doi.org/10.1101/2022.12.08.519575

Type: None

Postprocessing: None

Trained on: Accessible genomic sites in the mouse liver grouped into regulatory topics inferred from scATAC-seq data.

Source files

Specialized deep learning model to predict region accessibility (as topics) across cell types in the mouse liver.

Create a new conda environment with all dependencies installed
kipoi env create DeepLiver/DeepLiver_Accessibility
source activate kipoi-DeepLiver__DeepLiver_Accessibility
Copy
Test the model
kipoi test DeepLiver/DeepLiver_Accessibility --source=kipoi
Copy
Make a prediction
kipoi get-example DeepLiver/DeepLiver_Accessibility -o example
kipoi predict DeepLiver/DeepLiver_Accessibility \
  --dataloader_args='{"intervals_file": "example/intervals_file", "fasta_file": "example/fasta_file"}' \
  -o '/tmp/DeepLiver|DeepLiver_Accessibility.example_pred.tsv'
# check the results
head '/tmp/DeepLiver|DeepLiver_Accessibility.example_pred.tsv'
Copy
Create a new conda environment with all dependencies installed
kipoi env create DeepLiver/DeepLiver_Accessibility
source activate kipoi-DeepLiver__DeepLiver_Accessibility
Copy
Get the model
import kipoi
model = kipoi.get_model('DeepLiver/DeepLiver_Accessibility')
Copy
Make a prediction for example files
pred = model.pipeline.predict_example(batch_size=4)
Copy
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'])
Copy
Make predictions for custom files directly
pred = model.pipeline.predict(dl_kwargs, batch_size=4)
Copy
Get the model
library(reticulate)
kipoi <- import('kipoi')
model <- kipoi$get_model('DeepLiver/DeepLiver_Accessibility')
Copy
Make a prediction for example files
predictions <- model$pipeline$predict_example()
Copy
Use dataloader and model separately
# Download example dataloader kwargs
dl_kwargs <- model$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$inputs)
Copy
Make predictions for custom files directly
pred <- model$pipeline$predict(dl_kwargs, batch_size=4)
Copy
Get the docker image
Not available yet
Copy
Get the full sized docker image
Not available yet
Copy
Get the activated conda environment inside the container
Not available yet
Copy
Test the model
Not available yet
Copy
Make prediction for custom files directly
Not available yet
Copy
Install apptainer
https://apptainer.org/docs/user/main/quick_start.html#quick-installation-steps
Copy
Make prediction for custom files directly
Not available yet
Copy
Schema

Inputs

List of numpy arrays

Name: None

    Shape: (500, 4) 

    Doc: DNA sequence

Targets

Single numpy array

Name: topic

    Shape: (82,) 

    Doc: Topic Prediction

Dataloader

Defined as: .

Doc: Data-loader returning one-hot encoded sequences given genome intervals

Authors: Carmen Bravo

Type: None

License: MIT

Arguments

intervals_file : intervals file bed3

fasta_file : Reference genome FASTA file path.

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

Model dependencies
conda:
  • python=3.7
  • numpy==1.19.5
  • h5py==2.10.0
pip:
  • tensorflow==1.15.0
  • protobuf==3.20
Dataloader dependencies
conda:
  • python=3.7
  • bioconda::pybedtools
  • bioconda::pysam
  • bioconda::pyfaidx
  • numpy
  • pandas
pip:
  • kipoiseq