SSG-LUGIA: Single Sequence based Genome Level Unsupervised Genomic Island Prediction Algorithm
Background
Genomic Islands (GIs) are clusters of genes that are mobilized through horizontal gene transfer. GIs play a pivotal role in bacterial evolution as a mechanism of diversification and adaptation to different niches. Therefore, identification and characterization of GIs in bacterial genomes is important for understanding bacterial evolution. However, quantifying GIs is inherently difficult, and the existing methods suffer from low prediction accuracy and precision-recall trade-off. Moreover, several of them are supervised in nature and thus their applications to newly sequenced genomes are riddled with their dependency on the functional annotation of existing genomes.SSG-LUGIA
We present SSG-LUGIA, a completely automated and unsupervised approach for identifying GIs and horizontally transferred genes. SSG-LUGIA is a novel method based on unsupervised anomaly detection technique, accompanied by further refinement using cues from signal processing literature. SSG-LUGIA leverages the atypical compositional biases of the alien genes to localize GIs in prokaryotic genomes. SSG-LUGIA was assessed on the IslandPick benchmark dataset, and on the well-understood Salmonella typhi CT18 genome. Furthermore, the efficacy of SSG-LUGIA in identifying horizontally transferred genes was evaluated on two additional bacterial genomes, namely, that of Corynebacterium diphtheria NCTC13129 and Pseudomonas aeruginosa LESB58.Method
SSG-LUGIA takes an unannotated genome sequence as input and predicts GIs and optionally enlists putative
horizontally transferred
genes in case a gene annotation file is provided. The genome sequence of interest is first split into
overlapping
window frames that are then analyzed for the presence of GI distinguishing features compiled from
publicly
available literature on GIs. The genome sequence is screened for “anomalous” regions based on GI
distinguishing
features using an unsupervised machine learning procedure. The anomalous segments thus identified are
further
refined following a post-processing step, and finally, the proximal segments are merged to produce the
list
of GIs (and optionally, the list of horizontally transferred genes).
Conclusions
Our results indicate that SSG-LUGIA achieved superior performance in comparison to frequently used existing methods. Importantly, it yielded a better trade-off between precision and recall than the existing methods. Its non-dependency on the functional annotation of genomes makes it suitable for analyzing newly sequenced, yet uncharacterized genomes. Thus, our study is a significant advance in identification of GIs and horizontally transferred genes. The source code for SSG-LUGIA has been made open-source to aid the biologists in inferring genomic islands and horizontaly transferred genes from newly sequenced genomes. To further facilitate their discovery SSG-LUGIA will be made available as a web interface with a server application, along with portable cross-platform native applications in coming days.Source Code
The source code for SSG-LUGIA can be found in the following github repository.Usage
SSG-LUGIA provides a flexible way to use the pipeline as a command line interface
1. Clone the git repository.
2. Navigate to the /codes directory
3. Import the SSG-LUGIA pipeline from the main.py script.
4. Execute the SSG-LUGIA pipeline, using either a standard model (SSG-LUGIA-F, SSG-LUGIA-R, SSG-LUGIA-P) or a custom model configuration
$ git clone https://github.com/nibtehaz/SSG-LUGIA.git
2. Navigate to the /codes directory
3. Import the SSG-LUGIA pipeline from the main.py script.
4. Execute the SSG-LUGIA pipeline, using either a standard model (SSG-LUGIA-F, SSG-LUGIA-R, SSG-LUGIA-P) or a custom model configuration
(i) Input the path to the genome sequence and provide a standard model name
(ii) Input the path to the genome sequence and input the model parameters interactively
