Dataset Information

SAPPHIRE.CNN: Implementation of dRNA-seq-driven, species-specific promoter prediction using convolutional neural networks.

ABSTRACT: Data availability is a consistent bottleneck for the development of bacterial species-specific promoter prediction software. In this work we leverage genome-wide promoter datasets generated with dRNA-seq in the Gram-negative bacteria Pseudomonas aeruginosa and Salmonella enterica for promoter prediction. Convolutional neural networks are presented as an optimal architecture for model training and are further modified and tailored for promoter prediction. The resulting predictors reach high binary accuracies (95% and 94.9%) on test sets and outperform each other when predicting promoters in their associated species. SAPPHIRE.CNN is available online and can also be downloaded to run locally. Our results indicate a dependency of binary promoter classification on an organism's GC content and a decreased performance of our classifiers on genera they were not trained for, further supporting the need for dedicated, species-specific promoter classification tools.

SUBMITTER: Coppens L

PROVIDER: S-EPMC9478156 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

SAPPHIRE.CNN: Implementation of dRNA-seq-driven, species-specific promoter prediction using convolutional neural networks.

Coppens Lucas L Wicke Laura L Lavigne Rob R

Computational and structural biotechnology journal 20220909

Data availability is a consistent bottleneck for the development of bacterial species-specific promoter prediction software. In this work we leverage genome-wide promoter datasets generated with dRNA-seq in the Gram-negative bacteria <i>Pseudomonas aeruginosa</i> and <i>Salmonella enterica</i> for promoter prediction. Convolutional neural networks are presented as an optimal architecture for model training and are further modified and tailored for promoter prediction. The resulting predictors re ...[more]

PMID: 36147675

Similar Datasets

Project description:The prediction of breeding values and phenotypes is of central importance for both livestock and crop breeding. In this study, we analyze the use of artificial neural networks (ANN) and, in particular, local convolutional neural networks (LCNN) for genomic prediction, as a region-specific filter corresponds much better with our prior genetic knowledge on the genetic architecture of traits than traditional convolutional neural networks. Model performances are evaluated on a simulated maize data panel (n = 10,000; p = 34,595) and real Arabidopsis data (n = 2,039; p = 180,000) for a variety of traits based on their predictive ability. The baseline LCNN, containing one local convolutional layer (kernel size: 10) and two fully connected layers with 64 nodes each, is outperforming commonly proposed ANNs (multi layer perceptrons and convolutional neural networks) for basically all considered traits. For traits with high heritability and large training population as present in the simulated data, LCNN are even outperforming state-of-the-art methods like genomic best linear unbiased prediction (GBLUP), Bayesian models and extended GBLUP, indicated by an increase in predictive ability of up to 24%. However, for small training populations, these state-of-the-art methods outperform all considered ANNs. Nevertheless, the LCNN still outperforms all other considered ANNs by around 10%. Minor improvements to the tested baseline network architecture of the LCNN were obtained by increasing the kernel size and of reducing the stride, whereas the number of subsequent fully connected layers and their node sizes had neglectable impact. Although gains in predictive ability were obtained for large scale data sets by using LCNNs, the practical use of ANNs comes with additional problems, such as the need of genotyping all considered individuals, the lack of estimation of heritability and reliability. Furthermore, breeding values are additive by design, whereas ANN-based estimates are not. However, ANNs also comes with new opportunities, as networks can easily be extended to account for additional inputs (omics, weather etc.) and outputs (multi-trait models), and computing time increases linearly with the number of individuals. With advances in high-throughput phenotyping and cheaper genotyping, ANNs can become a valid alternative for genomic prediction.

Dataset Information

SAPPHIRE.CNN: Implementation of dRNA-seq-driven, species-specific promoter prediction using convolutional neural networks.

Publications

SAPPHIRE.CNN: Implementation of dRNA-seq-driven, species-specific promoter prediction using convolutional neural networks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets