The public SIEGA webpageSIEGA has a public website25 where a detailed description of the circuit and updated information on the species under surveillance is available for the general public. Having a public webpage for a project of genomic surveillance of pathogens is vital for promoting transparency, disseminating knowledge, fostering collaboration, and bridging the gap between the scientific community and the general public. By providing open access to information on the surveillance achievements, SIEGA encourages participation from diverse stakeholders, and creates a more knowledgeable and engaged society in the ongoing fight against infectious diseases. SIEGA offers access to Nextstrain Auspice26 viewers for the different species under surveillance: Salmonella enterica, Listeria monocytogenes, Campylobacter jejuni, Escherichia coli, Legionella pneumophila and Yersinia enterocolitica (see Fig. 2). In the viewers, the public can explore the relationships among the samples sequenced in the region and other international samples of reference. It is also possible to locate in a map the geographical origin of the different isolates, including the animated options available in the interface, which emulate in a very visual way the transmission of the samples over time.Figure 2Phylogenies from Nextstrain viewers for: (A) Salmonella enterica, (B) Listeria monocytogenes, (C) Campylobacter jejuni, (D) Escherichia coli, (E) Legionella pneumophila and (F) Yersinia enterocolitica.The SIEGA data management systemThe SIEGA data management system serves as a private LIMS designed for utilization by personnel in public health and participating clinicians of the surveillance program. SIEGA facilitates the seamless uploading of raw sequencing data and orchestrates automated processing, including quality control assessments, considering the Guidelines for reporting Whole Genome Sequencing-based typing data through the EFSA One Health WGS System of the European Food Safety Authority (EFSA)27. Through this platform, users can generate tailored reports concerning the isolates, explore potential transmission chains, and deploy an automated alert mechanism that promptly signals any genetic similarity between newly identified bacteria and entries within the existing database. This system bolsters vigilance and response capabilities. Furthermore, SIEGA furnishes statistical insights into the isolates, along with a comprehensive exploration of their interrelationships and the distribution of resistances across samples.Within the SIEGA interface, each organism has five distinct subsections to facilitate comprehensive analysis that include sample status, metadata, control results and Flexible Table, a wizard to combine metadata for complex representations (see Table 1).
Table 1 SIEGA features and description.The entire SIEGA data management system has been designed to free users of the need to have in-house expertise in genomic data management and resources to store such data. Simultaneously, the system offers a centralized database of all genomes sampled, allowing optimal exploitation of the results and the correct implementation of one health surveillance. It provides a convenient user permission structure to allows data sharing and collaborative work (if desired), detailed quality control for the standard pipelines used for data processing and accurate data traceability. Beyond facilitating collaborative work, user permission also helps with data privacy in samples of human origin. In any case, metadata does not include any field with personal identification and, ultimately, it is user responsibility not including any information that would reveal the origin of the sample. In general, data is treated in an open data philosophy where possible, according to FAIR principles (findable, accessible, interoperable and reusable)28.Detailed reports on each sample, that include MLST, cgMLST, serotyping outcomes, antimicrobial resistance and virulence genes, plasmids and a phylogenetic tree with the most related samples are provided. Additionally, customized phylogenetic analysis can be carried out in an easy and intuitive way. Finally, one of the most interesting features is the automatic alert system. SIEGA can be configured to automatically send a warning when a new sample is introduced that meets some criteria defined by the user, based on genetic distance, serotype, presence of antimicrobial or virulence genes, etc. (see Table 1 and Supplementary results for details).Advantages of SIEGAThe overarching goal of the SIEGA initiative, and particularly its SIEGA data management system, is to streamline the implementation of epidemiological surveillance, especially at the regional level or within large communities engaged in the One Health approach. It offers a continuously upgraded environment for managing genomic data, eliminating the need for end-users to establish their own bioinformatics teams for data processing and interpretation, as well as to invest in computer infrastructure for data storage and analysis. Furthermore, because data undergo uniform processing through a regularly updated pipelines adhering to international analysis standards, the results are consistent and can be readily compared. This democratizes genomic surveillance by involving all necessary stakeholders, as the SIEGA platform provides the essential resources for both data processing and interpretation.Current usersMicrobiology laboratories across different hospitals in Andalusia are actively utilizing SIEGA. In addition, the “Sistema de Vigilancia Epidemiológica de Andalucía” (SVEA) from the “Junta de Andalucía” and professionals in the field of Health Protection are also users of SIEGA. In adherence to the One Health principle, individuals involved in both animal health management and laboratory aspects have been incorporated as users of SIEGA.Prospective new users must contact SIEGA through the contact address in the public SIEGA web page.Genomic surveillance of isolates
Salmonella entericaThe SIEGA encompasses a dataset comprising 670 whole genome sequences of Salmonella enterica, which were sequenced from June 2020 to July 2023 using samples collected between 2013 and 2023. Within this dataset, 42.54% (285) of the samples were sourced from clinical origins, 34.63% from the food-related sector, and 21.34% from livestock sources. A total of 448 distinct Sequence Types (STs) were identified and categorized into clonal complexes (CCs). The prevailing ST, ST 309694 (corresponding to clonal complex ST-71), was encountered on 26 occasions. Additionally, 83 strains exhibited concurrence with more than one ST. 6 STs (ST-67337, ST-138467, ST-197094, ST-207307, ST-247937, and ST-320298) were found cross-wide clinical, food-related, and livestock-origin samples. Similarly, 18 STs were identified in both clinical and food samples, and 7 STs were identified in clinical and livestock-origin samples.
Listeria monocytogenesThe SIEGA includes a dataset comprising 678 whole genome sequences of Listeria monocytogenes, which were sequenced from June 2019 to July 2023. Within this dataset, 69.61% (472) of the samples were sourced from clinical origins, including all the samples from Andalusia sequenced by the Neisseria, Listeria and Bordetella Unit of the National Centre for Microbiology in Spain while investigating the Listeriosis outbreak caused by contaminated stuffed pork in Spain in 201929, and 30.38% (206) from food origin. A total of 248 distinct STs were identified and categorized into CCs. The prevailing ST, ST 29,514 (corresponding to clonal complex ST-388), was encountered on 210 occasions.Campylobacter sppThe SIEGA contains 276 whole genome sequences of Campylobacter, received between December 2020 and June 2023, corresponding to both C. jejuni and C. coli. Most of the sequences have been obtained from human clinical strains from two reference hospitals in Cádiz and Seville. There are some STs, grouped into CCs, which have been detected with greater frequency in clinical samples. From ST-16294 (corresponding to the clonal complex ST-206) 12 sequences have been obtained, with the interest of being detected from 2020 to 2023 and in a scattered way, with 8 isolates in Cádiz and 4 in Seville. Their identical virulome and resistome profiles have been recovered from these sequences, using the tools described, particularly ABRicate30 on VFDB31 and CARD32, databases. Another frequent STs have been ST-12550 (ST-573CC) and ST-18855 (ST-52CC).
Escherichia coliThe SIEGA includes 121 whole genome sequences of Escherichia coli, 44 of them downloaded from the EnteroBase website33 for reference and 77 of them collected between November 2021 and May 2023. To date, most of the sequences (72) have been obtained from food samples taken at retail level on behalf of the monitoring programme of anti-microbial resistance (AMR) according to the provisions of the Commission Implementing Decision (EU) 2020/172934 implemented in Andalusia. The human clinical strains come from two reference hospitals in Cádiz and Seville. There are STs, grouped into CCs, which have been detected with greater frequency in food samples. The prevailing ST, ST 169652 (corresponding to clonal complex ST-10) was encountered in 6 occasion all from food samples. Additionally, other 4 strains grouped in ST142026 (CC 155) and 3 strains exhibited concurrence with ST 191979 (CC 162) or 60064 (CC 93). To the date, no shared CCs have been detected in the food and human clinical origin samples.
Yersinia enterocoliticaThe SIEGA encompasses 23 whole genome sequences of Yersinia enterocolitica, received in the year 2022, sampled between February and November. To date 21 (91.3%) of the sequences have been obtained from clinical samples from one of the reference hospitals, the Hospital Virgen del Rocio in Seville, and 2 were obtained from food samples. There are some STs, grouped into CCs, which have been detected with greater frequency in these clinical samples. The prevailing ST, ST-1574 (corresponding to clonal complex ST-135), was encountered on 4 occasions. Additionally, 3 strains exhibited concurrence with ST-52 and 2 strains grouped into ST-1716 (corresponding also to clonal complex ST-135). Figure 3 depicts the genetic relationships between all the Yersinia enterocolitica samples.Figure 3Y. enterocolitica GrapeTree representation, generated within the SIEGA application. Node labels represent ST (in some cases an ambiguous ST assignation occurred and more than one number is displayed) and node color correspond to the sampling month (a warm gradient has been used to better display the time scale). Numbers in the branches correspond to the allelic distances among nodes. The GrapeTree representation provides an intuitive visualization of the temporal scale of sampling and the genetic similarities among the samples. Using different labels from the metadata and the results tables, it is possible to obtain visual representations of many aspects of the epidemiology of the selected samples.
Legionella pneumophilaLegionella data stored in SIEGA, includes the comparative analyses of 58 Legionella pneumophila isolates during 2021–2023. Of these, 12 isolates corresponded to clinical isolates and 46 to environmental isolates. Clonal relation between the Isolates was determined by cgMLST. This scheme classified the isolates into 18 ST (sequence type). The most abundant being ST 293 (20 isolates, 34.5%) and 180F (11 isolates, 18.9%). Four ST (293, 427, 489 and 180F) were present in both clinical and environmental isolates. In addition, we identified 3 STs (95, 98, and 524F) in clinical isolates that are not associated with environmental origin, suggesting that they derived from unrecognized sources.Analysis of antimicrobial resistanceIn the EU, the new legislation related to the harmonized monitoring and reporting of AMR from 202135 authorized whole genome sequencing as an alternative method to supplementary phenotypic testing of Salmonella and E. coli in certain conditions. The SIEGA allows the monitoring of the presence of resistance genes in the different microorganisms facilitating the tracking of the dissemination or emergence of AMR throughout the food chain under a One Health approach. For example, the presence of AMR genes in the population of Salmonella included in the SIEGA can be analyzed, categorized by antimicrobial classes (Fig. 4). This analysis reveals that 52.7% (347) exhibit resistance genes to only 1 group of antimicrobials, while 15% (99) carry resistance genes to two distinct classes of antimicrobials. In contrast, 32.2% (212) demonstrate resistance genes to 3 or more classes of antimicrobials. In a similar manner, this could be carried out with the other microorganisms hosted in the database, or further analysis could be conducted by delving into the multiple variables, for instance, this could involve monitoring the emergence of Salmonella strains harboring colistin resistance genes, the occurrence of Salmonella strains harboring resistance genes to fluoroquinolones and third-generation cephalosporins or monitoring the presence of resistance genes to Critically Important Antibiotics (CIAs) in the database. The Fig. 4 represents a summary of the observed frequency of potential multi-resistances, represented as the number of different AMR genes corresponding to different antimicrobial classes harbored by each individual sample.Figure 4Observed frequency of potential multi-resistance cases found among the Salmonella samples sequenced. Number of samples in which from only one to up to 9 different AMR genes have been found.Another perspective for monitoring antimicrobial resistance is through the surveillance of plasmids harboring these resistance genes. SIEGA flexible tables allow the integration of data to both antimicrobial resistance and plasmid tracking, facilitating a comprehensive analysis. This data can be graphically represented on a phylogenetic tree, which illustrates the STs that have acquired resistance-bearing plasmids. Moreover, this representation can highlight whether such acquisitions have occurred within the same time, geographical region, livestock farm, food processing plant, grocery store or healthcare facility, thus providing critical insights into the patterns and pathways of resistance spread. Figure 5 illustrates a case is the plasmid NZ_AJ437107, which harbors a beta-lactam resistance gene. This plasmid has been acquired in both livestock and food samples with the same sequence type, similar date and from the same geographical location, suggesting a potential selection pressure in certain environments favoring the acquisition of beta-lactam resistance genes.Figure 5Phylogenetic tree, based on allelic differences (log scale), of Salmonella enterica isolated from the same locality, grouped by allelic profile (the circle size correlates with the number of samples). Blue dots represent strains harboring the plasmid NZ_AJ437107, and the number inside each dot indicates the corresponding MLST for each sample.Some successful SIEGA use casesDespite its incipient use, SIEGA has already proven its usefulness in several cases. Among them it is worth mentioning the investigations carried out in connection with an outbreak of Salmonella Agona, as declared by Norwegian authorities36, entailed, irrespective of field investigations, a comprehensive review of the SIEGA database in pursuit of genomic congruences. This study resulted in the absence of any coincident strains. Another case was the investigation regarding the food alert notification issued under the identifier 2020.5961 within the European Commission Rapid Alert System for Food and Feed (RASFF)37, involving actions that extended beyond on-site measures. These actions included obtaining genomic sequences from food samples supplied by the Finnish food safety authorities. These sequences were then integrated into the SIEGA database. However, no matches were identified both at the time of integration and among subsequent samples added to the system. Further investigations undertaken in relation to the “Joint ECDC-EFSA rapid outbreak assessment” released on July 27, 202338, in which Spain was identified as one of the conceivable sources of the suspected transmission vehicle, involved, regardless of field actions, a reassessment of the SIEGA database in search of genomic coincidences, resulting in the non-existence of any matching strain.These cases clearly illustrate how SIEGA is valuable in ruling out the existence of coincident strains in our database with strains from evaluations of other outbreaks detected at the national or European level, thus facilitating decision-making. On the other hand, alerts are generated that point out possible connections between the stored strains and the ongoing outbreaks, accelerating the possible identification of the source of origin in an outbreak, an aspect that is very difficult to identify with previous methods.
The integrated genomic surveillance system of Andalusia (SIEGA) provides a One Health regional resource connected with the clinic
