Janggu - Deep learning for genomics

Motivation In recent years, numerous applications have demonstrated the potential of deep learning for an improved understanding of biological processes. However, most deep learning tools developed so far are designed to address a specific question on a fixed dataset and/or by a fixed model architecture. Adapting these models to integrate new datasets or to address different hypotheses can lead to considerable software engineering effort. To address this aspect we have built Janggu, a python library that facilitates deep learning for genomics applications. Results Janggu aims to ease data acquisition and model evaluation in multiple ways. Among its key features are special dataset objects, which form a unified and flexible data acquisition and pre-processing framework for genomics data that enables streamlining of future research applications through reusable components. Through a numpy-like interface, the dataset objects are directly compatible with popular deep learning libraries, including keras. Furthermore, Janggu offers the possibility to visualize predictions as genomic tracks or by exporting them to the BIGWIG format. We illustrate the functionality of Janggu on several deep learning genomics applications. First, we evaluate different model topologies for the task of predicting binding sites for the transcription factor JunD. Second, we demonstrate the framework on published models for predicting chromatin effects. Third, we show that promoter usage measured by CAGE can be predicted using DNase hyper-sensitivity, histone modifications and DNA sequence features. We improve the performance of these models due to a novel feature in Janggu that allows us to include high-order sequence features. We believe that Janggu will help to significantly reduce repetitive programming overhead for deep learning applications in genomics, while at the same time enabling computational biologists to assess biological hypotheses more rapidly. Availability Janggu is freely available under a GPL-v3 license on https://github.com/BIMSBbioinfo/janggu or via https://pypi.org/project/janggu

[1]  Ankur Taly,et al.  Axiomatic Attribution for Deep Networks , 2017, ICML.

[2]  Nathan C. Sheffield,et al.  The accessible chromatin landscape of the human genome , 2012, Nature.

[3]  B. Frey,et al.  Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning , 2015, Nature Biotechnology.

[4]  Annalisa Marsico,et al.  pysster: classification of biological sequences by learning sequence and structure motifs with convolutional neural networks , 2018, Bioinform..

[5]  Julian A. Peterson,et al.  A microcomputer network for biochemistry , 1985, Comput. Appl. Biosci..

[6]  O. Stegle,et al.  DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning , 2016, Genome Biology.

[7]  Jens Keilwagen,et al.  Varying levels of complexity in transcription factor binding motifs , 2015, Nucleic acids research.

[8]  Sanjiv Kumar,et al.  On the Convergence of Adam and Beyond , 2018 .

[9]  David R. Kelley,et al.  Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks , 2015, bioRxiv.

[10]  Fabian J Theis,et al.  Deep learning: new computational modelling techniques for genomics , 2019, Nature Reviews Genetics.

[11]  Yanjun Qi,et al.  DeepChrome: deep-learning for predicting gene expression from histone modifications , 2016, Bioinform..

[12]  Jun Cheng,et al.  Kipoi: accelerating the community exchange and reuse of predictive models for genomics , 2018, bioRxiv.

[13]  Xiaohui S. Xie,et al.  DanQ: a hybrid convolutional and recurrent deep neural network for quantifying the function of DNA sequences , 2015, bioRxiv.

[14]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[15]  David R. Kelley,et al.  Sequential regulatory activity prediction across chromosomes with convolutional neural networks. , 2018, Genome research.

[16]  O. Troyanskaya,et al.  Predicting effects of noncoding variants with deep learning–based sequence model , 2015, Nature Methods.

[17]  R. Mann,et al.  Quantitative modeling of transcription factor binding specificities using DNA shape , 2015, Proceedings of the National Academy of Sciences.

[18]  Anshul Kundaje,et al.  Discovering epistatic feature interactions from neural network models of regulatory DNA sequences , 2018, bioRxiv.

[19]  Patrice Y. Simard,et al.  Best practices for convolutional neural networks applied to visual document analysis , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[20]  O. Stegle,et al.  Deep learning for computational biology , 2016, Molecular systems biology.

[21]  Olga G. Troyanskaya,et al.  Selene: a PyTorch-based deep learning library for biological sequence-level data , 2018, bioRxiv.