Avsecz

# Where the dataloader is implemented and in what form
type: Dataset
defined_as: dataloader.py::SeqDataset

# Arguments of the dataloader
args:
    intervals_file:
        doc: tsv file containing dna interval indices (chr, start, end) and (optonally) binary 0/1 labels
        example: example_files/intervals_files.tsv
    fasta_file:

Avsecz

model = kipoi.get_model(
    "https://github.com/kipoi/models/tree/<commit>/<model>",
    source='github-permalink'
)

Avsecz

# figure size
# tilted xaxis
# from https://github.com/kipoi/manuscript/blob/master/src/transfer_learning/plot.ipynb

plotnine.options.figure_size = (5,2.5)
gplt = ggplot(aes(x='Cell_Type', y='auPRC', fill='Model'), data=dft) + \
    theme_classic() + \
    theme(axis_text_x=element_text(angle=20, hjust = 1)) + \
    theme(legend_title=element_blank(), 
          legend_box_margin=0,

Avsecz

import kipoi
from kipoi_interpret.importance_scores.gradient import GradientXInput

model = kipoi.get_model("DeepBind/Homo_sapiens/TF/D00765.001_ChIP-seq_GATA1")

val = GradientXInput(model).score(seq_array)[0]
seqlogo_heatmap(val, val.T)

Avsecz

# Create and activate a new conda environment
# with all model dependencies installed
kipoi env create <Model>
source activate kipoi-<Model>

# Run model predictions and save the results
# sequentially into an HDF5 file
kipoi predict <Model> --dataloader_args='{
  "intervals_file": "intervals.bed",
  "fasta_file": "hg38.fa"}' \

Avsecz

library(reticulate)
kipoi <- import('kipoi')
model <- kipoi$get_model('Basset')
model$predict_on_batch(x)

Avsecz

import kipoi

kipoi.list_models() # list available models

model = kipoi.get_model("Basset") # load the model

model = kipoi.get_model(  # load the model from a past commit
    "https://github.com/kipoi/models/tree/<commit>/<model>",
    source='github-permalink'
)

Avsecz

import numpy as np
import pandas as pd
from pybedtools import BedTool
from genomelake.extractors import FastaExtractor
from kipoi.data import Dataset
from kipoi.metadata import GenomicRanges


class SeqDataset(Dataset):

Avsecz

from torch.utils.data import DataLoader  
# if you don't want to install pytorch,
# you can use a fork in Kipoi:
# from kipoi.external.torch.data import DataLoader
from kipoi.data_utils import numpy_collate

ds = SeqDataset(fasta_file = '', ...)
dl = DataLoader(ds,
                batch_size=32,
                collate_fn=numpy_collate,

Avsecz

#!/bin/bash

# Build latest version of Emacs, version management with stow
# OS: Ubuntu 14.04 LTS
# version: 24.5
# Toolkit: lucid

# Warning, use updated version of this script in: https://github.com/favadi/build-emacs

set -e