Similarity Studies of Corona Viruses through Chaos Game Representation

Sengupta, Dipendra C.; Hill, Matthew D.; Benton, Kevin; Banerjee, Hirendra Nath

doi:10.4236/cmb.2020.103004

Cited by 9 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, the comparison between genomes using CGR is straightforward and fast ( Hoang et al , 2016 ). CGR has been used, for instance, for a fast comparison of SARS-CoV2 strains ( Sengupta et al , 2020 ). Extensions of CGR include color grids ( Deschavanne et al , 1999 ) and frequency matrix chaos game representation (FCGR) ( Almeida et al , 2001 ).…”

Section: Introductionmentioning

confidence: 99%

Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning

et al. 2021

View full text Add to dashboard Cite

Motivation Antimicrobial resistance (AMR) is one of the biggest global problems threatening human and animal health. Rapid and accurate AMR diagnostic methods are thus very urgently needed. However, traditional antimicrobial susceptibility testing (AST) is time-consuming, low throughput, and viable only for cultivable bacteria. Machine learning methods may pave the way for automated AMR prediction based on genomic data of the bacteria. However, comparing different machine learning methods for the prediction of AMR based on different encodings and whole-genome sequencing data without previously known knowledge remains to be done. Results In the current study, we evaluated logistic regression (LR), support vector machine (SVM), random forest (RF), and convolutional neural network (CNN) for the prediction of AMR for the antibiotics ciprofloxacin (CIP), cefotaxime (CTX), ceftazidime (CTZ), and gentamicin (GEN). We could demonstrate that these models can effectively predict AMR with label encoding, one-hot encoding, and frequency matrix chaos game representation (FCGR encoding) on whole-genome sequencing data. We trained these models on a large AMR dataset and evaluated them on an independent public data set. Generally, RFs and CNNs perform better than LR and SVM with AUCs up to 0.96. Furthermore, we were able to identify mutations that are associated with AMR for each antibiotic. Availability Source code in data preparation and model training are provided at GitHub website (https://github.com/YunxiaoRen/ML-iAMR). Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…For instance, multiple questions were addressed with this algorithm during the SARS-CoV-2 pandemic. For instance CGR and FCGR were used by Touati et al [61] and [62] , respectively, to build phylogenetic trees of coronaviruses. Sengupta et al [63] applied machine learning for taxonomic alignment-free classification of viruses, based on CGR.…”

Section: Conclusion and Future Perspectivesmentioning

confidence: 99%

Chaos game representation and its applications in bioinformatics

Löchel

Heider

2021

Computational and Structural Biotechnology Journal

View full text Add to dashboard Cite

show abstract

“…One of the most used feature extraction methods is the so-called FCGR, in which the CGR image is split into small grids and the frequencies of points falling into each grid are taken as the feature of the corresponding protein sequence. In our previous work [24], we used FCGR to study the similarity of coronavirus sequences. While FCGR has been used mainly for coronavirus genome sequence encoding and classification, we modified it in this study to work also for protein sequences.…”

Section: Introductionmentioning

confidence: 99%

Evolutionary Relationship of Protein Sequences of SARS-CoV-2 and Other Viruses through Chaos Game Representation

Hill¹,

Simmons²,

Sengupta³

2022

CMB

View full text Add to dashboard Cite

Comparison between different biological sequences is a key step in bioinformatics when analyzing similarities of sequences and phylogenetic relationships. A method of graphically representing biological sequences known as Chaos Game Representation (CGR) has achieved many applications in the studies of bioinformatics. The key issue in the application of CGR is to extract as many useful features as possible from CGR. Initially, CGR was applied to DNA sequences, but in this paper, a CGR-based approach is used to extract suitable features for comparing protein sequences of SARS-CoV-2 and other viruses. For this aim, several viral protein sequences from 12 groups are considered and CGR centroid, amino acid frequency, compounded frequency, Shannon entropy, and Kullback-Lieber Discrimination Information are applied to find the inter-relationship among the sequences. The experimental results demonstrate the potential strengths of CGR-based method for examining the evolutionary relationship of protein sequences. Our method is powerful for extracting effective features from protein sequences, and therefore important in classifying proteins and inferring the phylogeny of viruses.

show abstract

Similarity Studies of Corona Viruses through Chaos Game Representation

Cited by 9 publications

References 17 publications

Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning

Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning

Chaos game representation and its applications in bioinformatics

Evolutionary Relationship of Protein Sequences of SARS-CoV-2 and Other Viruses through Chaos Game Representation

Contact Info

Product

Resources

About