Julian A Paganini

Dr. Julian A Paganini


Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2020 in Microbial Genomics
Verified
Journal Article
Open access
MCMaría M. Cameranesi
María M. Cameranesi
JPJulian A Paganini
Julian A Paganini
ALAdriana S. Limansky
Adriana S. Limansky
JMJorgelina Moran-Barrio
Jorgelina Moran-Barrio
SSSuzana P. Salcedo
Suzana P. Salcedo
GRGuillermo D. Repizo
Guillermo D. Repizo
Acinetobacter baumannii (Aba) is an emerging opportunistic pathogen associated to nosocomial infections. The rapid increase in multidrug resistance (MDR) among Aba strains underscores the urgency of understanding how this pathogen evolves in the clinical environment. We conducted here a whole-genome sequence comparative analysis of three phylogenetically and epidemiologically related MDR Aba strains from Argentinean hospitals, assigned to the CC104O/CC15P clonal complex. While the Ab244 strain was carbapenem-susceptible, Ab242 and Ab825, isolated after the introduction of carbapenem therapy, displayed resistance to these last resource β-lactams. We found a high chromosomal synteny among the three strains, but significant differences at their accessory genomes. Most importantly, carbapenem resistance in Ab242 and Ab825 was attributed to the acquisition of a Rep_3 family plasmid carrying a bla OXA-58 gene. Other differences involved a genomic island carrying resistance to toxic compounds and a Tn10 element exclusive to Ab244 and Ab825, respectively. Also remarkably, 44 insertion sequences (ISs) were uncovered in Ab825, in contrast with the 14 and 11 detected in Ab242 and Ab244, respectively. Moreover, Ab825 showed a higher killing capacity as compared to the other two strains in the Galleria mellonella infection model. A search for virulence and persistence determinants indicated the loss or IS-mediated interruption of genes encoding many surface-exposed macromolecules in Ab825, suggesting that these events are responsible for its higher relative virulence. The comparative genomic analyses of the CC104O/CC15P strains conducted here revealed the contribution of acquired mobile genetic elements such as ISs and plasmids to the adaptation of A. baumannii to the clinical setting.

16 January 2025 in Microorganisms
Verified
Journal Article
Open access
PDF file
CBCamila Maria dos Santos Boralli
Camila Maria dos Santos Boralli
JPJulian A Paganini
Julian A Paganini
RMRodrigo Silva Meneses
Rodrigo Silva Meneses
CMCamila Pacheco Silveira Martins da Mata
Camila Pacheco Silveira Martins da Mata
ELEdna Marilea Meireles Leite
Edna Marilea Meireles Leite
Ilana L. B. C. CamargoIlana L. B. C. Camargo
Ilana L. B. C. Camargo
KPC is a clinically significant serine carbapenemase in most countries, and its rapid spread threatens global public health. blaKPC transmission is commonly mediated by Tn4401 transposons. The blaKPC gene has also been found in non-Tn4401 elements (NTEKPC). To fill the gap in the understanding of the stability and dissemination of NTEKPC-carrying plasmids, we selected and characterized carbapenem-resistant bacteria isolated between 2009 and 2016 from a hospital for a retrospective study of their plasmids conjugation capacity, impact on fitness, and replication in different species. Different clones were selected using PFGE, and their genomes were sequenced using Illumina and Oxford Nanopore methods. Minimum inhibitory concentrations (MICs) were determined by broth microdilution. Plasmid copy numbers (PCNs) were determined using qPCR. Doubling time was used to analyze fitness change. Most isolates (67%, 33/49) carried blaKPC, of which 85% presented blaKPC in a NTEKPC. The 25 isolates selected presented the blaKPC gene in NTEKPC-IId in IncQ1-type plasmids, showing multispecies dissemination. IncQ1 plasmids were mobilizable and PCN seemed to be directly linked to the species, presenting a high-copy number, mainly in K. pneumoniae. No relationship was observed between IncQ1 PCN and carbapenems MIC values. IncQ1 and a conjugative plasmid from K. pneumoniae BHKPC10 were transferred to E. coli J53 without fitness changes, and MIC values were maintained for carbapenems despite the low transconjugant PCN. In addition to IncQ1 with NTEKPC, Enterobacter cloacae BHKPC28 contained the mcr-9 gene in an IncHI2/IncHI2A conjugative plasmid, which may help the mobility of IncQ1 and the dissemination of two resistance determinants to last-resort antibiotics. Understanding the interaction between plasmids and high-risk lineages can help develop new therapies to prevent the dissemination of resistance traits.

7 July 2021 in bioRxiv
Verified
Preprint
Open access
PDF file
JPJulian A Paganini
Julian A Paganini
NPN.L. Plantinga
N.L. Plantinga
SAS. Arredondo-Alonso
S. Arredondo-Alonso
RWR.J.L. Willems
R.J.L. Willems
ASAnita C Schürch
Anita C Schürch
The incidence of infections caused by multidrug-resistant E. coli strains has risen in the past years. Antibiotic resistance in E. coli is often mediated by acquisition and maintenance of plasmids. The study of E. coli plasmid epidemiology and genomics often requires long-read sequencing information, but recently a number of tools that allow plasmid prediction from short-read data have been developed. Here, we reviewed 25 available plasmid prediction tools and categorized them into binary plasmid/chromosome classification tools and plasmid reconstruction tools. We benchmarked six tools that aim to reliably reconstruct distinct plasmids, with a special focus on plasmids carrying antibiotic resistance genes (ARGs) such as extended-spectrum beta-lactamase genes. They use either assembly graph information (plasmidSPAdes, gplas), reference databases (MOB-Suite, FishingForPlasmids) or both (HyAsP and SCAPP) to produce plasmid predictions. The benchmark data set consisted of 240 E. coli strains, harboring 631 plasmids, which were representative for the diversity of E. coli in public databases. Notably, these strains were not used for training any of the tools. We found that two thirds (n=425, 66.3.%) of all plasmids were correctly reconstructed by at least one of the six tools, with a range of 92 (14.58%) to 317 (50.23%) correctly predicted plasmids. However, the majority of plasmids that carried antibiotic resistance genes (n=85, 57.8%) could not be completely recovered as distinct plasmids by any of the tools. MOB-suite was the only tool that was able to correctly reconstruct the majority of plasmids (n=317, 50.23%), and performed best at reconstructing large plasmids (n=166, 46.37%) and ARG-plasmids (n=41, 27.9%), but predictions frequently contained chromosome contamination (40%). In contrast, plasmidSPAdes reconstructed the highest fraction of plasmids smaller than 18 kbp (n=168, 61.54%). Large ARG-plasmids, however, were recovered with small precision values (median=0.47, IQR=0.61), indicating that plasmidSPAdes frequently merged sequences derived from distinct replicons. Additionally, only 63% of all plasmid-borne ARGs were correctly predicted by plasmidSPAdes. The remaining four tools (FishingForPlasmids, HyAsP, SCAPP and gplas) were able to correctly reconstruct a combined total of 18 plasmids that were missed by MOB-suite and plasmidSPAdes. Available bioinformatic tools can provide valuable insight into E. coli plasmids, but also have important limitations. This work will serve as a guideline for selecting the most appropriate plasmid reconstruction tool for studies focusing on E. coli plasmids in the absence of long-read sequencing data.

17 December 2025 in Microbial Genomics
Verified
Journal Article
Open access
JPJulian A Paganini
Julian A Paganini
SKSuniya Khatun
Suniya Khatun
SMSean McAteer
Sean McAteer
LCLauren Cowley
Lauren Cowley
DGDavid R. Greig
David R. Greig
TDTimothy J. Dallman
Timothy J. Dallman
Shiga toxin-producing Escherichia coli (STEC) O157:H7 is a globally dispersed zoonotic pathogen capable of causing severe disease outcomes, including bloody diarrhoea and haemolytic uraemic syndrome. While variations in Shiga toxin subtype are well-recognized drivers of disease severity, many unexplained differences remain among strains carrying the same toxin profile. We applied explainable machine learning (ML) approaches – Random Forest and Extreme Gradient Boosting – to whole-genome sequencing data from 1,030 STEC O157:H7 isolates to predict patient clinical outcomes, using data collected over 2 years of routine surveillance in England. A phylogeny-informed cross-validation strategy was implemented to account for population structure and avoid data leakage, ensuring robust model generalizability. Extreme Gradient Boosting outperformed Random Forest in predicting minority classes and correctly predicted high-risk isolates in traditionally low-risk lineages, illustrating its utility for capturing complex genomic signatures beyond known virulence genes. Feature importance analyses highlighted phage-encoded elements, including potentially novel intergenic regulators, alongside established virulence factors. Moreover, key genomic regions linked to small RNAs and stress-response pathways were enriched in isolates causing severe disease. These findings underscore the capacity of explainable ML to refine risk assessments, offering a valuable tool for early detection of high-risk STEC O157:H7 and guiding targeted public health interventions.

3 September 2023 in bioRxiv
Verified
Preprint
Open access
PDF file
JPJulian A Paganini
Julian A Paganini
JKJesse J. Kerkvliet
Jesse J. Kerkvliet
LVLisa Vader
Lisa Vader
NPNienke L. Plantinga
Nienke L. Plantinga
RMRodrigo Meneses
Rodrigo Meneses
ASAnita C Schürch
Anita C Schürch
Accurate reconstruction of Escherichia coli antibiotic resistance gene (ARG) plasmids from Illumina sequencing data has proven to be a challenge with current bioinformatic tools. In this work, we present an improved method to reconstruct E. coli plasmids using short reads. We developed plasmidEC, an ensemble classifier that identifies plasmid-derived contigs by combining the output of three different binary classification tools. We showed that plasmidEC is especially suited to classify contigs derived from ARG plasmids with a high recall of 0.941. Additionally, we optimised gplas, a graph-based tool that bins plasmid-predicted contigs into distinct plasmid predictions. Gplas2 is more effective at recovering plasmids with large sequencing coverage variations and can be combined with the output of any binary classifier. The combination of plasmidEC with gplas2 showed a high completeness (median=0.818) and F1-score (median=0.812) when reconstructing ARG plasmids and exceeded the binning capacity of the reference-based method MOB-suite. In the absence of long read data, our method offers an excellent alternative to reconstruct ARG plasmids in E. coli . Data Summary: No new sequencing data have been generated in this study. All genomes used in this research are publicly available at the GenBank and Sequence Read Archive of the National Center for Biotechnology Information. Accession numbers are specified in Supplementary Materials. Scripts to reproduce the results reported in this manuscript can be accessed at https://gitlab.com/jpaganini/ecoli-binary-classifier . The ensemble classifier, plasmidEC, is publicly available at https://gitlab.com/mmb-umcu/plasmidEC (release 1.3.1), and gplas2 (release 1.0.0) can be found at https://gitlab.com/mmb-umcu/gplas2 . Impact Statement: Escherichia coli has emerged as a highly pervasive multidrug resistant pathogen on a global scale. The dissemination of resistance is significantly influenced by plasmids, mobile genetic elements that facilitate the transfer of antimicrobial resistance genes within and between diverse bacterial species. Consequently, precise and high-throughput identification of plasmids is imperative for effective genomic surveillance of resistance. However, accurate plasmid reconstruction remains challenging with the use of affordable short-read sequencing data. In this work, we present a novel method to accurately predict and reconstruct E. coli plasmids based on Illumina data. Additionally, we demonstrate that our approach outperforms the reference-based method MOB-suite, especially when reconstructing plasmids carrying antimicrobial resistance genes.

6 June 2025 in medRxiv
Verified
Preprint
Open access
PDF file
JPJulian A Paganini
Julian A Paganini
SKSuniya Khatun
Suniya Khatun
SMSean P McAteer
Sean P McAteer
LCLauren Cowley
Lauren Cowley
DGDavid R. Greig
David R. Greig
TDTimothy J. Dallman
Timothy J. Dallman
Background: Shiga toxin-producing Escherichia coli (STEC) O157:H7 is a globally dispersed zoonotic pathogen capable of causing severe disease outcomes, including bloody diarrhoea and haemolytic uraemic syndrome. While variations in Shiga toxin subtype are well-recognised drivers of disease severity, many unexplained differences remain among strains carrying the same toxin profile. Results: We applied explainable machine learning approaches—Random Forest and Extreme Gradient Boosting—to whole-genome sequencing data from 1,030 STEC O157:H7 isolates to predict patient clinical outcomes, using data collected over two years of routine surveillance in England. A phylogeny-informed cross-validation strategy was implemented to account for population structure and avoid data leakage, ensuring robust model generalizability. Extreme Gradient Boosting outperformed Random Forest in predicting minority classes and correctly predicted high-risk isolates in traditionally low-risk lineages, illustrating its utility for capturing complex genomic signatures beyond known virulence genes. Feature importance analyses highlighted phage-encoded elements, including potentially novel intergenic regulators, alongside established virulence factors. Moreover, key genomic regions linked to small RNAs and stress-response pathways were enriched in isolates causing severe disease. Conclusions: These findings underscore the capacity of explainable ML to refine risk assessments, offering a valuable tool for early detection of high-risk STEC O157:H7 and guiding targeted public health interventions.