LOGISTAR — Scheda Oggetto

Dati oggetto

URL

Apri

Titolo / Descrizione

Sito / Azienda

Estratto / Body

&lt;p&gt;&amp;nbsp;Deep learning (DL) has become a popular tool to study cis-regulatory element function. Yet efforts&lt;br /&gt;to design software for DL analyses in genomics that are Findable, Accessible, Interoperable and&lt;br /&gt;Reusable (FAIR) have fallen short of fully meeting these criteria.&lt;/p&gt;
&lt;p&gt;Here we present EUGENe&lt;br /&gt;(Elucidating the Utility of Genomic Elements with Neural Nets), a FAIR toolkit for the analysis of&lt;br /&gt;labeled sets of nucleotide sequences with DL. EUGENe consists of a set of modules that&lt;br /&gt;empower users to execute the key functionality of a DL workflow:&lt;/p&gt;
&lt;p&gt;1) extracting, transforming and&lt;br /&gt;loading sequence data from many common file formats,&lt;/p&gt;
&lt;p&gt;2) instantiating, initializing and training&lt;br /&gt;diverse model architectures, and&lt;/p&gt;
&lt;p&gt;3) evaluating and interpreting model behavior.&lt;/p&gt;
&lt;p&gt;We designed EUGENe to be simple; users can develop and deploy workflows on new or existing datasets with&lt;br /&gt;functions that act on two customizable Python objects, annotated sequence data (SeqData) and&lt;br /&gt;PyTorch models (BaseModel).&lt;/p&gt;
&lt;p&gt;The modularity and simplicity of EUGENe also make it highly&lt;br /&gt;extensible and we illustrate these principles through application of the toolkit to three predictive&lt;br /&gt;modeling tasks. First, we train and compare a set of built-in models along with a custom&lt;br /&gt;architecture for the accurate prediction of activities of plant promoters from STARR-seq data.&lt;br /&gt;Next, we apply EUGENe to an RNA binding prediction task and showcase how seminal model&lt;br /&gt;architectures can be retrained in EUGENe or imported from Kipoi. Finally, we train models to&lt;br /&gt;classify transcription factor binding by wrapping functionality from Janngu, which can efficiently&lt;br /&gt;extract sequences in BED file format from the human genome. We emphasize that the code used&lt;br /&gt;in each use case is simple, readable, and well documented (https://eugene-&lt;br /&gt;tools.readthedocs.io/en/latest/index.html).&lt;/p&gt;
&lt;p&gt;We believe that EUGENe represents a springboard&lt;br /&gt;toward a collaborative ecosystem for DL applications in genomics research. EUGENe is available&lt;br /&gt;for download on GitHub (https://github.com/cartercompbio/EUGENe) along with several&lt;br /&gt;introductory tutorials and for installation on PyPi (https://pypi.org/proje&lt;/p&gt;

Note

Voto oggetto (1-10)

Precisione (1-10)

Tag (5)

✎ Modifica tag 🤖 Suggerisci AI

           genoma,tool,python         

gen 5 genoma 5 python 5 tool 5 wpg 5

ID: 355 • Tipo: WPG • Reg: 2023-10-09 • Mod: 10/11/2026 23:13