Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

dnadna: DEEP NEURAL ARCHITECTURES FOR DNA - A DEEP LEARNING FRAMEWORK FOR POPULATION GENETIC INFERENCE

Théophile Sanchez 1 Erik Madison Bray 1 Pierre Jobic 1 Jérémy Guez 1, 2 Guillaume Charpiat 1 Jean Cury 1, * Flora Jay 3, 4, 1, *
* Corresponding author
3 TAU - TAckling the Underspecified
Inria Saclay - Ile de France, LISN - Laboratoire Interdisciplinaire des Sciences du Numérique
4 BioInfo - BioInformatique
LISN - Laboratoire Interdisciplinaire des Sciences du Numérique, SDD - Science des Données
Abstract : We present dnadna, a flexible python-based software for deep learning inference in population genetics. It is task-agnostic and aims at facilitating the development, reproducibility, dissemination, and reusability of neural networks designed for genetic polymorphism data. dnadna defines multiple user-friendly workflows. First, users can implement new architectures and tasks, while benefiting from dnadna input/output and other utility functions, training procedure and test environment, which not only saves time but also decreases the probability of bugs. Second, implemented networks can be re-optimized based on user-specified training sets and/or tasks. Finally, users can apply pretrained networks in order to predict evolutionary history from alternative real or simulated genetic datasets, without the need of extensive knowledge in deep learning. Thanks to dnadna, newly implemented architectures and pretrained networks are easily shareable with the community for further benchmarking or applications. dnadna comes with a peer-reviewed exchangeable neural network allowing demographic inference from SNP data, that can be used directly or retrained to solve other tasks. Toy networks are also available to ease the exploration of the software, and we expect that the range of available architectures will keep expanding thanks to contributions from the community.
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-03352910
Contributor : Flora Jay Connect in order to contact the contributor
Submitted on : Thursday, September 23, 2021 - 3:56:29 PM
Last modification on : Wednesday, December 1, 2021 - 3:34:13 AM

File

DNADNA_software_arxiv_submissi...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-03352910, version 1

Collections

Citation

Théophile Sanchez, Erik Madison Bray, Pierre Jobic, Jérémy Guez, Guillaume Charpiat, et al.. dnadna: DEEP NEURAL ARCHITECTURES FOR DNA - A DEEP LEARNING FRAMEWORK FOR POPULATION GENETIC INFERENCE. 2021. ⟨hal-03352910v1⟩

Share

Metrics

Record views

120

Files downloads

141