Whole genome dataset of bacterial strains isolated from urine analysed by Oxford Nanopore sequencing technologies in Ouagadougou, Burkina Faso

Abstract

A collection of 21 bacterial isolates recovered from urine specimens collected between September 2018 and February 2019 at Saint-Camille and SCHIPHRA hospitals in Ouagadougou, Burkina Faso, is presented. Isolates were identified using the API 20E system and tested for antibiotic susceptibility according to CASFM 2018 guidelines. Genomic DNA was extracted, barcoded with the SQK-RBK114.96 Rapid Barcoding Kit, and sequenced on an Oxford Nanopore MinION Mk1C using an R10.4.1 flowcell. Basecalling was performed with Dorado, and reads were quality filtered prior to de novo assembly with Flye, followed by polishing with Racon and Medaka. Assemblies were evaluated with QUAST, taxonomically assigned with Kraken2, and annotated with Bakta. The dataset comprises 21 barcode-specific raw read sets (1.5 million reads; 4.9 Gb), polished genome assemblies, per-isolate metadata including collection site, collection date, and antibiotic susceptibility measurements, and quality assessment outputs including read statistics, assembly metrics, and functional annotation completeness estimates. All data are available under NCBI BioProject PRJNA1307828. These data provide a useful resource for long-read bacterial genome assembly benchmarking, comparative genomics, and analyses integrating phenotypic and genomic antimicrobial resistance information.

Specifications Table

Subject	Molecular Biology
Specific subject area	Genomics, Microbiology, Bioinformatics
Type of data	Raw sequencing reads (FASTQ) and assembled genomes (FASTA)
Data collection	Bacterial strains were isolated from urine from out-patients. The cultures were incubated at 37 °C for 24 h. Isolated colonies were cultured in Luria Bertani (LB) broth to reach a density of 0.5 nm. Bacterial genomic DNA was extracted using the Qiagen QIAamp DNA kit. Whole genome sequencing was performed on Oxord Nanopore sequencing technology using SQK-RBK114.96 kit. Basecalling was performed using dorado (v.0.9.1), and raw reads were cleaned to remove contamination using minimap2 (v.2.28). then De novo assembly was performed by Flye (v.2.95) software, contigs were polished using medaka (v2.0.1) and racon (v1.4.20) and their quality was assessed by quast (5.2.0), taxonomique assignment were done using kraken2 (v2.0.8). Genome annotation was performed with the Bakta pipeline (v1.12.0)
Data source location	Geographical coordinates where the sample was collected: SCHIPHRA 12°23′27″N 1°31′00″W, St Camille Hospital 12°21′59″N 1°29′42"W District/Province: Kadiogo, Ouagadougou Country: Burkina Faso
Data accessibility	Repository name: National Center for Biotechnology Information (NCBI) Data identification number: BioProject: PRJNA1307828, BioSample: from SAMN50676038 to SAMN50676058, Sequence Read Archive (SRA): SRR35031978, SRR35031977, SRR35031966, SRR35031964, SRR35031963, SRR35031962, SRR35031961, SRR35031960, SRR35031959, SRR35031958, SRR35031976, SRR35031975, SRR35031974, SRR35031973, SRR35031972, SRR35031971, SRR35031970, SRR35031969, SRR35031968, SRR35031967, SRR35031965. Direct URL to data: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1307828
Related research article	None

1. Value of the Data

• •The genomic sequence data from Burkina Faso’s patient fills genomic data gap in West Africa and will be useful for researchers working in the different bacterial genomic fields.
• •These data will allow researchers to validate antibiotic resistance profiles and identify potential candidate genes associated with resistance.
• •These bacterial genomic data can serve as reference resources for epidemiological studies and surveillance.
• •It supports comparative genomic and antimicrobial resistance studies.
• •It enables the creation of training datasets for bioinformatics and microbial genomics capacity building in West Africa.

2. Background

Bacterial infections continue to represent a significant challenge to public health, particularly in settings where the capacity for microbiological surveillance and the characterisation of resistance mechanisms is constrained [[1], [2], [3]]. It is mentioned that unsuccessful treatment of bacterial infections associated with antibiotic resistance claims at least 700,000 every year globally and is projected to be associated with 10 million deaths per year by 2050 [4]⁠ In Burkina Faso, urinary tract and suppurative infections are prevalent and are frequently attributable to multidrug-resistant Enterobacteriaceae or non-fermenting bacilli. [5]. The establishment of a local genomic resource documenting these pathogens was achieved through the sequencing of clinical isolates obtained from urine samples collected at Saint Camille Hospital and SCHIPHRA Hospital (Ouagadougou).

The generation of genomic data was accomplished through the utilisation of Oxford Nanopore MinION long-read sequencing, which was followed by a standard assembly workflow (Flye, Racon, Medaka) and functional annotation with Bakta. For each isolate, the dataset comprises raw reads, assembled genomes, quality control reports and key clinical metadata (sample type, healthcare facility, collection date and antibiotic susceptibility profile). The present data article delineates the structure of the collection and the experimental protocol, thus providing a reference framework for investigations into bacterial resistance in West Africa.

3. Data Description

This dataset includes raw FASTQ sequencing data and de novo assembled whole-genome sequences (contigs in FASTA format) for clinical bacterial isolates obtained from urine samples, as well as associated metadata (Table 1). These isolates were collected between September 2018 and February 2019 from Saint Camille and SCHIPHRA hospitals in Ouagadougou, Burkina Faso. They were then stored at −80 °C until analysis.

Table 1.

Clinical bacterial isolates analysed in this study, with sample metadata, genome accession numbers, and antimicrobial resistance profiles.

Sample ID	Species	Sex	Age (year)	Hospital	Collection period	SRA Accession number	Genome assembly Accession number	Resistance profile
BCEc-5	Enterobacter hormaechei	M	53	SCHIPHRA	9/20/2019	SRR35031978	JBRVIB000000000	AML, CRO, CAZ, FOS
BCKp-97	Bacillus subtilis	M	17	SCHIPHRA	8/30/2019	SRR35031977	JBRVIV000000000	AML, AUG, CRO, CFX, SXT, CIP, OFX
BCEc-23	Escherichia coli	M	75	SCHIPHRA	1/7/2019	SRR35031966	JBRVIU000000000	AML, SXT, CIP, OFX
BCEc-U3	E. coli	F	42	Saint-Camille	9/28/2018	SRR35031964	JBRVIT000000000	AML, AUG, CRO, CFM, SXT, GN, TOB, CIP, OFX
BCEc-2977	E. coli	F	49	SCHIPHRA	2/12/2019	SRR35031963	JBRVIS000000000	AML, CRO, CFM, SXT, CIP, OFX
BCEc-U4	E. coli	M	76	Saint-Camille	1/11/2019	SRR35031962	JBRVIR000000000	AML, CRO, CFM, CIP, OFX
BCEc-91	E. coli	M	72	SCHIPHRA	1/10/2019	SRR35031961	JBRVIQ000000000	AML, CRO, CFM, SXT, CIP, OFX
BCEc-U18	E. coli	F	0.25	Saint-Camille	1/18/2019	SRR35031960	JBRVIP000000000	AML, AUG, CRO, CFM, SXT
BCEc-U5	E. coli	M	60	Saint-Camille	11/30/2018	SRR35031959	JBRVIO000000000	AML, CRO, CFM, CIP, OFX
BCEc-120	E. coli	F	22	SCHIPHRA	1/17/2019	SRR35031958	JBRVIN000000000	AML, CRO, CFM, SXT, CIP, OFX
BCEc-U1	E. coli	M	NA	Saint-Camille	1/18/2019	SRR35031976	JBRVIM000000000	AML, CRO, CFM, SXT, CIP, OFX
BCEc-U9-1	E. coli	F	60	Saint-Camille	9/28/2018	SRR35031975	JBRVIL000000000	AML, AUG, CRO, CFM, SXT, GN, TOB, CIP, OFX
BCEc-U7	E. coli	F	02	Saint-Camille	12/7/2018	SRR35031974	JBRVIK000000000	AML, AUG, CRO, CFM, SXT, GN, TOB, CIP, OFX
BCEc-U9-2	E. coli	F	37	Saint-Camille	1/18/2019	SRR35031973	JBRVIJ000000000	AML, AUG, CRO, CFM, SXT, TOB, CIP, OFX
BCEc-U11	E. coli	M	23	Saint-Camille	12/7/2018	SRR35031972	JBRVII000000000	AML, AUG, CRO, CFM, SXT, AK, TOB, CIP, OFX
BCEc-U14-1	E. coli	NA	NA	Saint-Camille	11/9/2018	SRR35031971	JBRVIH000000000	AML, AUG, CRO, CFM, SXT, TOB, CIP, OFX
Ec-2844	E. coli	M	72	SCHIPHRA	2/10/2019	SRR35031970	JBRVIG000000000	AML, AUG, CRO, CFM, SXT, TOB, CIP, OFX
BCEc-U4-19	E. coli	M	66	Saint-Camille	1/18/2019	SRR35031969	JBRVIF000000000	AML, AUG, CRO, CFM, SXT, TOB, K, CIP, OFX
BCEc-2939	E. coli	M	60	SCHIPHRA	2/12/2019	SRR35031968	JBRVIE000000000	AML, AUG, CRO, CFM, SXT, GN, CIP, OFX
BCEc-U14-2	E. coli	F	NA	Saint-Camille	11/23/2018	SRR35031967	JBRVID000000000	AML, AUG, CRO, CFM, SXT, TOB, CIP, OFX
BCEc-U10	E. coli	F	62	Saint-Camille	9/28/2018	SRR35031965	JBRVIC000000000	AML, AUG, CRO, CFM, SXT, GN, TOB, CIP, OFX

NA: Not available. F: Female. M: Male. Antibiotics tested: Amoxicillin (AML), Amoxicillin + Clavulanic acid (AUG), Imipenem (IMI), Ceftriaxone (CRO), Cefixime (CFM), Ceftazidime (CAZ), Trimethoprim + Sulfamethoxazole (SXT), Amikacin (AK), Gentamicin (GN), Kanamycin (K), Tobramycin (TOB), Neomycin (N), Netilmicin (NET), Ciprofloxacin (CIP) Ofloxacin (OFX), Fosfomycin (FOS).

Genomic DNA was extracted from the bacterial cells and sequencing was performed using the Rapid Barcoding Kit 96 V14 on a MinION sequencer (Oxford Nanopore Technologies, UK). Raw nanopore signal data were basecalled using Dorado (v0.9.1) with the super accuracy (SUP) model (dna_r10.4.1_e8.2_400bps_sup@v5.0.0). The resulting data were then converted to FASTQ format for downstream analysis. De novo genome assemblies were generated using the Flye v2.9.5 assembler (Fig. 1& 2). Circular genome representation annotated isolates assembly generated using Bakta annotation. It illustrates the overall genome architecture, including CDS distribution, RNA features and compositional profiles, such as GC content and GC skew (Fig. 3).

Fig. 1.

Per-barcode distribution of read counts and mean sequencing quality.

Fig. 2.

Comparison of assembly quality across samples as a function of sequencing coverage and read yield.

Fig. 3.

Genome annotation of representative bacterial isolates obtained after assembly.

Both the raw reads and the corresponding assembled genomes have been deposited in the NCBI Sequence Read Archive (SRA) under BioProject accession PRJNA1307828 (Table 1). This dataset provides valuable genomic resources for studying the genetic diversity, antimicrobial resistance and epidemiology of circulating clinical bacteria in Burkina Faso.

A total of 1.3 million Nanopore reads were generated across all barcodes, with read counts ranging from 11,061 to 170,995. The mean read length varied from 4141 bp to 8024 bp, with an average of around 6500 bp overall. Basecalling quality scores were consistent across samples, with mean Q-scores above 24 indicating reliable Nanopore sequencing performance (Table 2, Fig 2).

Table 2.

Nanopore read metrics and Flye assembly statistics for the 21 barcoded clinical isolates.

Sample ID	SRA Accession number	Genome assembly Accession number	Reads (n)	Mean read length (bp)	Mean Q (≥ 20)	Contig length (bp)	Depth (×)	Read-length N50 (bp)	Bases sequenced (bp)
BCEc‑5	SRR35031978	JBRVIB000000000	14 448	8 024	25.63	4 651 798	24	13 588	50 719 827
BCKp‑97	SRR35031977	JBRVIV000000000	11 778	6 741	25.05	4 112 813	17	9 718	35 236 226
BCEc‑23	SRR35031966	JBRVIU000000000	64 597	5 584	24.91	5 109 248	65	8 891	215 311 369
BCEc‑U3	SRR35031964	JBRVIT000000000	30 151	7 007	25.47	5 078 894	40	11 614	93 769 648
BCEc‑2977	SRR35031963	JBRVIS000000000	68 985	7 269	25.40	4 873 199	99	11 014	309 220 386
BCEc‑U4	SRR35031962	JBRVIR000000000	81 369	5 525	25.19	5 169 818	80	8 583	282 848 723
BCEc‑91	SRR35031961	JBRVIQ000000000	66 842	4 466	24.64	5 104 163	49	7 499	193 840 917
BCEc‑U18	SRR35031960	JBRVIP000000000	52 266	6 185	25.53	4 932 147	63	8 955	194 703 759
BCEc‑U5	SRR35031959	JBRVIO000000000	137 212	5 948	25.07	4 862 359	167	9 279	394 297 515
BCEc‑120	SRR35031958	JBRVIN000000000	13 951	4 141	23.45	2 290 274	12	5 014	40 690 756
BCEc‑U1	SRR35031976	JBRVIM000000000	170 995	6 192	24.93	4 698 334	225	9 107	666 036 325
BCEc‑U9‑1	SRR35031975	JBRVIL000000000	85 029	5 290	24.75	4 683 128	88	8 355	196 148 055
BCEc‑U7	SRR35031974	JBRVIK000000000	63 322	6 986	25.62	4 948 569	87	11 513	187 347 005
BCEc‑U9‑2	SRR35031973	JBRVIJ000000000	24 488	7 063	25.50	4 819 458	33	10 543	103 296 199
BCEc‑U11	SRR35031972	JBRVII000000000	24 077	6 188	25.32	4 984 733	29	9 047	94 840 476
BCEc‑U14‑1	SRR35031971	JBRVIH000000000	134 315	4 605	24.53	4 768 506	75	6 685	248 182 873
Ec‑2844	SRR35031970	JBRVIG000000000	58 583	7 435	25.43	4 980 102	83	11 402	251 090 135
BCEc‑U4‑19	SRR35031969	JBRVIF000000000	65 714	7 148	25.09	4 891 137	29	11 641	265 097 121
BCEc‑2939	SRR35031968	JBRVIE000000000	65 714	7 148	25.09	4 845 335	52	11 641	265 097 121
BCEc‑U14‑2	SRR35031967	JBRVID000000000	55 098	7 364	24.97	4 653 154	86	12 084	167 622 397
BCEc‑U10	SRR35031965	JBRVIC000000000	98 623	6 532	24.86	4 653 154	138	10 475	271 383 799

Reads (n): number of quality-filtered reads. Depth (×): average coverage of the longest contig.

Barcode a and b: two contigs of large size.

Genome assemblies produced one or a few large contigs per isolate, with total genome sizes ranging from ∼2.3 to ∼5.2 Mb, which is consistent with the expected size of bacterial genomes (Fig 1& 2). The longest contigs were up to 5.17 Mb in length, and assembly depths ranged from 12 × to 225 ×, ensuring sufficient coverage for accurate consensus generation. Read length N50 values ranged from 5014 bp to 13,588 bp, confirming the predominance of long reads that are well suited to high-quality assembly (Table 2, Fig 1).

Overall, the sequencing effort produced >4.3 billion bases, providing deep coverage (Fig 2) and robust assemblies that are suitable for downstream comparative genomic and phylogenetic analyses. These assemblies serve as reference-quality genomes for the clinical bacterial isolates sequenced in this study.

All sequencing reads and assemblies have been deposited in the NCBI Sequence Read Archive under BioProject PRJNA1307828. Collectively, these files furnish a consistent resource for the assessment of assembly pipelines, the examination of genome structure, and the correlation of genotypic characteristics to observed susceptibility profiles in clinically pertinent Escherichia coli, Bacillus subtilis, Enterobacter hormaechei, and associated pathogens from Burkina Faso and West Africa.

4. Experimental Design, Materials and Methods

4.1. Sample collection, bacterial culture and bacterial isolation

The study was conducted on 21 bacterial strains collected between September 2018 and February 2019 in biomedical laboratories of Saint Camille and SCHIPHRA hospitals in Ouagadougou. The 21 strains included in this dataset were selected from an initial collection of isolates based on DNA quality and suitability for sequencing. These isolates correspond to non-duplicate bacteria isolates originating from urine cultures of outpatients undergoing routine diagnostic testing. Samples were obtained from two hospitals (Saint Camille and SCHIPHRA). The samples were inoculated on a chromogenic medium, Chromatic Detection from Liofilchem, and incubated at 37 °C for 24 h to promote bacterial growth. Strain identification was confirmed using the API 20E gallery (Biomérieux). An antibiogram was then performed according to the CASFM 2018 guidelines (Table 1). The susceptibility of the strains to antibiotics was determined using the disk diffusion method (Kirby-Bauer), allowing the measurement of inhibition diameters and establishing the resistance profile to commonly used antibiotics.

For E. coli, Bacillus, and Enterobacter, the tested antibiotics were:

β-lactams: Amoxicillin (AML, 30 µg), Amoxicillin + Clavulanic acid (AUG, 20µg/10 µg), and Imipenem (IMI, 10 µg), Third-generation cephalosporins (C3G): Ceftriaxone (CRO, 30 µg), Cefixime (CFM, 5 µg), Ceftazidime (CAZ, 30 µg), Sulfonamides: Trimethoprim + Sulfamethoxazole (SXT, 1.25µg/23.75 µg), Aminoglycosides: Amikacin (AK, 30 µg), Gentamicin (GN, 10 µg), Kanamycin (K, 30 µg), Tobramycin (TOB, 10 µg), Neomycin (N, 30 µg), and Netilmicin (NET, 10 µg), Other: Fosfomycin (200 µg).

4.2. DNA extraction and whole genome sequencing

Bacterial strains were isolated from urine specimens collected from outpatients. After collection, each clinical specimen was inoculated onto Luria-Bertani (LB) agar and incubated at 37 °C for 24 h. Single colonies were then grown in LB broth until they reached an optical density at 600 nm (OD_600) of 0.5, as measured by spectrophotometry.

Genomic DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The concentration and purity of the extracted DNA were assessed by spectrophotometry (NanoDrop™ 2000, Thermo Fisher Scientific) and fluorometry (Qubit™ 4, Thermo Fisher Scientific) before library preparation.

Whole genome sequencing (WGS) was performed on an Oxford Nanopore Technologies (ONT) platform using the SQK-RBK114.96 Rapid Barcoding Kit (Oxford Nanopore Technologies, UK). Libraries were prepared according to the manufacturer's recommendations and loaded onto an R10.4.1 flow cell. Sequencing was performed until adequate coverage was achieved (typically >30 × per strain). Base calling of the raw signal (POD5 files) to nucleotide sequences (FASTQ) was performed using Dorado (v 0.9.1) (Oxford Nanopore Technologies).

4.3. Bioinformatics analysis

Raw reads were further processed to remove low-quality and contaminating sequences. This involved aligning the reads using Minimap2 (v 2.28) [7] ⁠using the recommended parameters for Nanopore long reads. De novo assembly was then performed using Flye (v 2.95) [8], which is designed to handle long and potentially error-prone reads.

Subsequent polishing steps were implemented to correct residual errors in the assemblies:

• -Medaka (v 2.0.1) (Oxford Nanopore Technologies) was first used to remove base-level errors specific to nanopore reads.
• -Racon (v 1.4.20) [3] was then used to further refine the consensus sequence through iterative alignment and polishing rounds.

The quality and integrity of the final assemblies were assessed using QUAST (v 5.2.0) [6], which generated metrics such as total assembly size, number of contigs, and completeness estimates. Kraken2 (v2.0.8) was used as a rapid, initial taxonomic classifier to determine the species with the highest level of confidence [9], using updated bacterial databases to ensure accurate classification.

The assembled genomes of the 21 bacterial isolates were annotated in terms of both structure and function using Bakta v1.12.0 with the full v6.0 database. Bakta was chosen because it rapidly and comprehensively annotates bacterial genomes by detecting coding and non-coding genomic elements. Bakta predicted coding sequences (CDSs) and annotated major RNA features for each genome assembly, including tRNAs, rRNAs, ncRNAs, and regulatory non-coding RNA regions. The pipeline also screened for additional genomic features, such as CRISPR regions and putative replication- and mobility-related loci (including oriC and oriT), when present. Functional assignments were inferred using Bakta's curated reference database and built-in annotation workflow. The annotation outputs were subsequently used for genome description, comparative interpretation and circular genome visualization (Fig 3). All annotations were generated using Bakta v1.12.0 and database v6.0 (full), with the standard parameters recommended by the developers being followed.

All bioinformatic analyses were performed in the BurkinaBioinfo (BBi) platform [10], using default parameters recommended by the tool’s developers, unless otherwise specified.

Limitations

Not applicable.

Ethics Statement

The Institutional Ethics Committee of Saint Camille Hospital of Ouagadougou (HOSCO), by Deliberation No. 2018-09-016 dated September 6, 2018, approved the research protocol of this study.

Credit Author Statement

Pamane Djagbare: Methodology, Validation, Formal analysis, Writing – Review and Editing. Ezechiel Bionimian Tibiri: Conceptualization, Methodology, Software, Validation, Formal analysis, Supervision, Data curation, Writing – Original Draft. Wendyam Marie Christelle Nadembèga: Conceptualization, Methodology, Data Curation, Writing – Review and Editing, supervision, Project administration, Resources, Funding acquisition. Pakyendou Estel Name: Methodology, Validation, Formal analysis, Visualization, Writing – Review and Editing. Lassina Traoré: Methodology, Validation, Formal analysis, Writing – Review and Editing. Emmanuel Sampo: Writing – Review and Editing. Moussa Ouédraogo: Writing – Review and Editing. Fidèle Tiendrebéogo: Methodology, Validation, Formal analysis, Investigation, Writing – Review and Editing, Resources. Jacques Simporé: Validation, Investigation, Writing – Review and Editing, Resources.

Data availability

• Raw sequencing data in FASTQ and FASTA formats have been deposited in the NCBI database under BioProject accession number PRJNA1307828. Due to their size, the raw signal data in POD5 format are not hosted in a public repository but are available from the corresponding author upon reasonable request.