1 Introduction
Along with the steady increase of multi-resistant and extensively virulent microorganisms, the genomic approach has become an essential ally in the search for genetic factors related to microbial pathogenicity (Mbelle et al., 2019). Thus, the call for tools capable of handling a large scale of information in a short period of time has increased (Suárez-Díaz, 2010).
The fact is that genomic information tends to be difficult to interpret due to the high information density present in several datasets (Nusrat et al., 2019). Moreover, the visualization of genomic analysis results becomes more complex whenever new genomes are added to the initial dataset and additional information is computed.
PanViTa (Pan Virulence and resisTance Analysis) was developed with the concepts of scalability and agility in mind. It is a tool made entirely in Python3 (Van Rossum and Drake, 2009), focusing on the analysis of multi-omic bacterial data based on complete or draft genomes. This tool was initially designed to handle data annotated by the PROKKA pipeline (Seemann, 2014) using GenBank files as input (.gbk or.gbff). However, it has been adapted to receive any GenBank file—with some reservations.
The tool is available on GitHub through the link https://github.com/dlnrodrigues/panvita.
2 Materials and methods
2.1 Implementation
The tool uses databases to obtain biological information available through the web, including CARD (Comprehensive Antimicrobial Resistance Database) (Alcock et al., 2020) and BacMet2 (Antibacterial Biocide and Metal Resistance Genes Database) (Pal et al., 2014) for resistance analysis, and VFDB (Virulence Factor Database) (Liu et al., 2019) for virulence analysis. The user can select any of the databases initially via the command line. BLASTp (Altschul et al., 1990; Mount, 2007) was selected in conjunction with the DIAMOND algorithm to compare the user data with the database reference (Buchfink et al., 2015).
For some features of the developed tool, it was necessary to take advantage of some existing libraries and modules in the native language. Therefore, the use of the Python3 version is recommended. Besides intrinsic modules and libraries (sys, OS, shutil, and math), the program also needs to import other modules and libraries: wget, to get external data and update databases and dependencies whenever necessary; pandas (Reback et al., 2021) for matrix construction and manipulation; seaborn (Waskom et al., 2020) and matplotlib (Hunter, 2007) both for the final plotting of graphical results. PanViTa requires 17 Mb of hard disk space for installation.
To obtain the final result, the program performs the following steps: (I) extracts the amino acid sequences of the predicted proteome from each GenBank file; (II) extracts the positions of the coding sequences of all proteins in the genome from each GenBank file; (III) aligns the predicted proteome with the selected database using DIAMOND-BLASTp; (IV) filters the results in the tabular output file that match the identity and coverage parameters (by default, results above 70% identity and 70% coverage are considered); (V) summarizes the results in a similarity-based matrix: X represents the genes with a match higher than the defined cutoff and Y represents the strains given as input; (VI) uses the summarized results to generate a clustermap plot based on Euclidean distance to determine data clusters by proximity; (VII) plots the development of each subpartition in core- and pan-; (VIII) checks and summarizes the gene results by specific strain. The flow chart containing the software steps is shown in Supplementary Figure S1.
Figure 1 represents some of the outputs generated using the PanViTa tool.
FIGURE 1
2.2 Comparative analysis
The performance of PanViTa was compared with other tools developed with a similar purpose: Abricate (Seemann, 2020) and ResFinder (Bortolaia et al., 2020). For comparison, genomes from different Acinetobacter baumannii, Escherichia coli, and Pseudomonas aeruginosa were selected. Table 1 represents the results obtained. The assembly codes from the strains of A. baumannii, E. coli, and P. aeruginosa are presented on Supplementary Table S1.
TABLE 1
| Acinetobacter baumannii | Escherichia coli | Pseudomonas aeruginosa | ||
|---|---|---|---|---|
| Number of genomes | 100 | 65 | 50 | |
| Genome size (Mb) ≅ | 4.03 | 5.24 | 7.15 | |
| Time | PanViTa | 00:02:08 | 00:01:36 | 00:01:23 |
| Abricate | 00:04:34 | 00:04:29 | 00:08:59 | |
| ResFinder | 01:13:52 | 01:06:26 | 01:06:13 | |
| Number of genes | PanViTa | 145 | 112 | 146 |
| Abricate | 119 | 120 | 119 | |
| ResFinder | 99 | 65 | 91 | |
Results obtained based on the comparison of the three tools analyzed.
For standardization purposes only the CARD database was added to the comparison analysis with Abricate.
The analyses were made on a desktop computer with OS Linux-Ubuntu viewing 8 GB of RAM and four cores (Intel® Core i5-3570 CPU 3.40 GHz).
3 Comparative results
Regarding the time spent, PanViTa was superior to the other tools compared. This fact can be related to both DIAMOND and BLASTp, which increase the alignment speed. The analysis of resistance genes is relative and dependent on the database, not being subject to direct comparison. Another important fact is the difference between the alignment matrices since PanViTa uses the amino acid sequences as the primary input, and Abricate and ResFinder consider nucleotide sequences. Besides, only PanViTa generates visual output.
The specific results obtained for each species are available in Supplementary Table S2.
4 Outputs
PanViTa provides some results based on the presence/absence of genes. Through this methodology it is possible to swiftly extract quantitative information about the action mechanisms of certain gene products, as well as which compounds are related to them.
4.1 Presence/absence matrix
One of the main outputs is a presence/absence matrix for each database, containings all identity values for each gene in each strain. All values are retrieved from multiple alignments against the previous selected database. Only the highest identity values per gene are considered for matrix building.
4.2 Clustermap
Euclidean distance is used as the metric to plot the clustermaps. In this way, it is possible to identify which genes are statistically related to each other. In addition, this data also enables to infer which strains are more or less related to each other using only a few resistance or pathogenicity genes presence.
4.3 Strain-specific genes
These both outputs are related to presence/absence statistics. With the usage of these outputs, it is possible to obtain the number and families of genes found on each bacterial strain properly, as well as the number of strains that share the same gene.
4.4 Virulence and resistance factors
PanViTa generates a single file for each strain containg the positions of CDSs related to specific virulence and resistance factors found on previous analysis. This file keeps the information from the original. gbk file, otherwise, if the original genome is not complete and has multiple contigs, the positions will be consecutive and addictive. In other words, if there’s more than one contig on GenBank’s file, to the positions extracted from consecutives contigs will be added the value of the length of the previous contigs.
4.5 Pan-ome curve
The pan-omic curve is an approximation of the pan-genome curve obtained from a basic pangenomic analysis. It is important to note, however, that this output has small statistical power because it is a plot of gene distribution from both sections of the pan-genomic approach—core genome and accessory genome. Non-etheless, it is interesting to observe, for example, if the pan-ome curve reached a stable point. Otherwise, the dataset considered for the analysis has a chance to continue to get over new resistance or virulence factors.
4.6 Antibiotics
When the selected database is CARD or BacMet2, it is possible to obtain a table that quantifies genes related to each antibiotic class obtained, being grouped by sub partition along the pan-resistome (central, accessory and exclusive). In this way, it is possible to assess the presence of certain target factors in more specific portions of the sample.
4.7 Other functions
PanViTa also has a genome acquisition module that allows the download of genomes available on the NCBI platform. For this, it is only necessary to use the. csv file generated during the genome search as input. In addition, it is possible to automatically generate the script for annotation using the prokka pipeline, and obtain the host and related disease metadata using the available biosample number.
Statements
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.
Author contributions
FA and VA conceived the idea. DR promoted data curation, formal analysis and methodology. DR and JA wrote the original draft. FR, DB, FA, and AMB-I revised and edited the original draft. VA and FA supervised the development of the project. VA and FA performed the funding acquisition. DR developed the final data visualization. All authors have read and agreed to the published version of the manuscript.
Funding
The work was financially supported by the Coordination for the Improvement of Higher Education Personnel (CAPES), the Minas Gerais State Research Support Foundation (FAPEMIG), the National Council for Scientific and Technological Development (CNPq) and the Pró-Reitoria de Pesquisa (PRPq) - UFMG.
Acknowledgments
Our thanks to the Post Graduate Program in Bioinformatics at the Federal University of Minas Gerais. We would also like to thank the laboratories associated with the Omics Science Network (RECOM), as well as the fomentation agencies.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fbinf.2023.1070406/full#supplementary-material
References
1
AlcockB. P.RaphenyaA. R.LauT. T. Y.TsangK. K.BouchardM.EdalatmandA.et al (2020). Card 2020: Antibiotic resistome surveillance with the comprehensive antibiotic resistance database. Nucleic Acids Res.48, D517–D525. 10.1093/nar/gkz935
2
AltschulS. F.GishW.MillerW.MyersE. W.LipmanD. J. (1990). Basic local alignment search tool. J. Mol. Biol.215, 403–410. 10.1016/S0022-2836(05)80360-2
3
BortolaiaV.KaasR. S.RuppeE.RobertsM. C.SchwarzS.CattoirV.et al (2020). ResFinder 4.0 for predictions of phenotypes from genotypes. J. Antimicrob. Chemother.75, 3491–3500. 10.1093/jac/dkaa345
4
BuchfinkB.XieC.HusonD. H. (2015). Fast and sensitive protein alignment using DIAMOND. Nat. Methods12, 59–60. 10.1038/nmeth.3176
5
HunterJ. D. (2007). Matplotlib: A 2D graphics environment. Comput. Sci. Eng.9, 90–95. 10.1109/MCSE.2007.55
6
LiuB.ZhengD.JinQ.ChenL.YangJ. (2019). Vfdb 2019: A comparative pathogenomic platform with an interactive web interface. Nucleic Acids Res.47, D687–D692. 10.1093/nar/gky1080
7
MbelleN. M.FeldmanC.Osei SekyereJ.ManingiN. E.ModipaneL.EssackS. Y. (2019). The resistome, mobilome, virulome and phylogenomics of multidrug-resistant Escherichia coli clinical isolates from pretoria, south Africa. Sci. Rep.9, 16457. 10.1038/s41598-019-52859-2
8
MountD. W. (2007). Using the basic local alignment search tool (BLAST). Cold Spring Harb. Protoc.2007, pdb.top17. 10.1101/pdb.top17
9
NusratS.HarbigT.GehlenborgN. (2019). Tasks, techniques, and tools for genomic data visualization. Comput. Graph. Forum J. Eur. Assoc. Comput. Graph.38, 781–805. 10.1111/cgf.13727
10
PalC.Bengtsson-PalmeJ.RensingC.KristianssonE.LarssonD. G. J. (2014). BacMet: Antibacterial biocide and metal resistance genes database. Nucleic Acids Res.42, D737–D743. 10.1093/nar/gkt1252
11
RebackJ.McKinneyW.jbrockmendelJ.AugspurgerT.CloudP.gfyoungS H.et al (2021). pandas-dev/pandas Pandas1. 10.5281/zenodo.4524629
12
SeemannT. (2014). Prokka: Rapid prokaryotic genome annotation. Bioinformatics30, 2068–2069. 10.1093/bioinformatics/btu153
13
SeemannT. (2020). tseemann/abricate.
14
Suárez-DíazE. (2010). Making room for new faces: Evolution, genomics and the growth of bioinformatics. Hist. Philos. Life Sci.32, 65–89.
15
Van RossumG.DrakeF. L. (2009). Python 3 reference manual. Scotts Valley, CA: CreateSpace.
16
WaskomM.GelbartM.BotvinnikO.OstblomJ.PaulH.LukauskasS.et al2020. mwaskom/seaborn: v0. 11. 1(December 2020). 10.5281/zenodo.4379347
Summary
Keywords
bioinformatics tools, data visualization, multiomics analysis, pathogens, resistome
Citation
Rodrigues DLN, Ariute JC, Rodrigues da Costa FM, Benko-Iseppon AM, Barh D, Azevedo V and Aburjaile F (2023) PanViTa: Pan Virulence and resisTance analysis. Front. Bioinform. 3:1070406. doi: 10.3389/fbinf.2023.1070406
Received
14 October 2022
Accepted
27 January 2023
Published
07 February 2023
Volume
3 - 2023
Edited by
Arne Seitz, Swiss Federal Institute of Technology Lausanne, Switzerland
Reviewed by
Leandro de Mattos Pereira, University of Porto, Portugal
Naouel Klibi, Tunis El Manar University, Tunisia
Updates
Copyright
© 2023 Rodrigues, Ariute, Rodrigues da Costa, Benko-Iseppon, Barh, Azevedo and Aburjaile.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Flávia Aburjaile, faburjaile@gmail.com
†These authors have contributed equally to this work
This article was submitted to Integrative Bioinformatics, a section of the journal Frontiers in Bioinformatics
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.