Kuangyi Charles Wei

Kuangyi Charles Wei

Auto-generated profile, based on publicly available information

Kuangyi Charles Wei

Department of Genetics, University of Cambridge, Cambridge, United Kingdom, CB2 3EH

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
8 August 2025 in Microbes and Infection
Verified
Journal Article
Open access
KWKuangyi Charles Wei
Kuangyi Charles Wei
SPSrinithi Purushothaman
Srinithi Purushothaman
FAFrancesca Azzato
Francesca Azzato
KBKate S. Baker
Kate S. Baker
KBKira Waagner Birkeland
Kira Waagner Birkeland
AEAdrian Egli
Adrian Egli

2 March 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
KWKuangyi (Charles) Wei
Kuangyi (Charles) Wei
BBBeth Blane
Beth Blane
JTJacqueline Toussaint
Jacqueline Toussaint
SRSandra Reuter
Sandra Reuter
MTMichelle S. Toleman
Michelle S. Toleman
WRWilliam Roberts-Sengier
William Roberts-Sengier
Microbial genome-wide association study (GWAS) tools often require manual data processing steps, lack comprehensive workflows, and are limited by scalability issues, thus hindering the exploration of bacterial genetic traits. To address these challenges, we developed BaGPipe, an automated and flexible bacterial GWAS pipeline built using Nextflow and incorporating Pyseer for association analysis. BaGPipe integrates all essential components of a bacterial GWAS—spanning pre-processing, statistical analysis, and downstream visualisation—into a unified workflow that is reproducible and easy to deploy across diverse computational environments. BaGPipe was validated on a publicly available dataset of Streptococcus pneumoniae whole-genome sequences, and reproduced published findings with improved computational efficiency. BaGPipe was then applied to a dataset of Staphylococcus aureus whole-genome sequences, successfully identifying known and novel antibiotic resistance associations. By offering an accessible, efficient, and reproducible platform, BaGPipe accelerates bacterial GWAS and facilitates deeper exploration into the genetic underpinnings of phenotypic traits. Impact Statement: The increasing availability of bacterial genome sequences has created an opportunity for robust, reproducible tools to facilitate the discovery of novel genotype-phenotype associations. Despite the demonstrated utility of genome-wide association studies (GWAS) in identifying genetic determinants of disease, toxicity and antibiotic resistance, existing tools for bacterial GWAS often involve fragmented workflows requiring extensive manual intervention, limiting their adoption and reproducibility. Here, we introduce BaGPipe, a fully integrated bacterial GWAS pipeline that automates pre-processing, statistical analysis, and visualisation, thereby streamlining the entire workflow. With its flexibility, scalability, and ease of use, BaGPipe makes bacterial GWAS more accessible to researchers, enabling faster and more reliable insights into microbial genetics. This is an important step towards overcoming the computational and logistical barriers that have constrained bacterial GWAS, ultimately accelerating research into microbial evolution, resistance mechanisms, and the genetic basis of other key phenotypic traits. Data Summary: BaGPipe is freely available at https://github.com/sanger-pathogens/BaGPipe . The Streptococcus pneumoniae input dataset is available from the Pyseer tutorial ( https://pyseer.readthedocs.io/en/master/tutorial.html# ). The Staphylococcus aureus sequencing assemblies can be sourced from their ERS accession numbers provided in supplementary data. The reference assemblies, listed in the supplementary, can be sourced from NCBI.

12 May 2026 in medRxiv
Verified
Preprint
Open access
PDF file
KWKuangyi Charles Wei
Kuangyi Charles Wei
CCCharlotte E. Chong
Charlotte E. Chong
GBGherard Batisti Biffignandi
Gherard Batisti Biffignandi
LMLewis C. E. Mason
Lewis C. E. Mason
RMRyan Morrison
Ryan Morrison
KBKate Baker
Kate Baker
Shigella sonnei is a human-adapted enteric pathogen with a very low infectious dose and increasing antimicrobial resistance. In high-income settings, transmission is multimodal including sporadic cases/outbreaks associated with food and travel, as well as sustained transmission among sexual networks of men who have sex with men (MSM). Whole-genome sequencing (WGS) now underpins national shigellosis surveillance in the United Kingdom. Hence, consistent, communicable genotyping is essential for case linkage and trend detection across heterogeneous transmission modes. Here, we evaluate the performance of WGS genotyping approaches for granulating outbreaks of S. sonnei shigellosis, particularly considering differential performance in dense sexual transmission where highly clonal MSM-associated sublineages pose distinct clustering challenges. Specifically, we compare performance of the current practice approach (10 SNP-distance clustering based on SNP address [t10]), allele-based methods (EnteroBase cgMLST/HierCC [HC5]), a pathogen-specific genotyping scheme (sonneityper), and two k-mer based approaches (PopPUNK and KPop), on a bona fide UK surveillance dataset (n = 3,639 isolates from between 2016 and 2022), and stratify analyses by demographics (i.e. presumptive MSM [pMSM] versus non-pMSM). Comparison metrics indicate that t10 clustering method groups data more broadly than HC5, and k-mer-based methods may capture genetic variation independent from SNP or allele-based approaches. Clusters derived from k-mer-based methods offer similar resolution to HC5 and reflect different demographics, but had unconvincing utility for this pathogen. These findings suggest a transmission context-aware surveillance workflow for shigellosis in high income settings: anchor routine communication on a portable allele-based backbone and augment with more granular, complementary methods (e.g., k-mer-based micro-partitioning or phylogenetic analysis) in comparatively low genomic-density regions of population structure (e.g., pMSM transmission lineages) to stabilise clusters and reduce artefactual chaining.

13 January 2026 in The Journal of Infectious Diseases
Verified
Journal Article
Closed access
SCSumanth Cherukumilli
Sumanth Cherukumilli
AKAdama M Keita
Adama M Keita
WSWilliam Still
William Still
KWKuangyi Charles Wei
Kuangyi Charles Wei
JJJane Juma
Jane Juma
KKKaren L Kotloff
Karen L Kotloff
Background: Despite data demonstrating the high prevalence of third-generation cephalosporin resistant Gram-negative Enterobacterales (GNE) among children in sub-Saharan Africa, longitudinal descriptions of resistance among these pathogens from Francophone West Africa remain rare. Methods: We conducted a retrospective analysis of pathogen-positive children 0–15 years old included in an invasive bacterial infection study at l’Hôpital Gabriel Touré (in Bamako, Mali) from 2005 to 2023. We aimed to describe changes in pathogen burden and non-Salmonella GNE resistance over time and compare in-hospital mortality between patients with and without non-Salmonella GNE pathogens. Isolates from 2021 to 2023 underwent whole-genome sequencing to identify genes conferring antimicrobial resistance. Results: Of 3803 pathogen-positive patients, 392 grew at least 1 non-Salmonella GNE pathogen. The proportion of pathogen-positive patients with non-Salmonella GNE increased (6% vs 38%) from 2005–2023. Third-generation cephalosporin and multidrug resistance among non-Salmonella GNE increased from 30% and 55%, respectively, in 2005–2009, to 93% (both) by 2021–2023. Children 0–2 months old from outside Bamako and 3 months-15 years old from and from outside Bamako with non-Salmonella GNE had higher mortality odds (3.17, 95% CI 1.69–5.95; 2.13, 95% CI 1.44–3.14; and 2.25, 95% CI 1.38–3.66, respectively) than similar patients with other pathogens. Sequencing confirmed the presence of the emerging pathogen Pantoea dispersa and revealed genes conferring multidrug resistance. Conclusions: Data from this large pediatric referral center in Mali show a high and rising burden of multidrug-resistant Gram-negative Enterobacterales. These patterns reflect concerns increasingly reported across sub-Saharan Africa, highlighting the urgency of strengthening antimicrobial access, diagnostics, and stewardship strategies in similar settings.

14 September 2026 in medRxiv
Verified
Preprint
Open access
PDF file
GBGherard Batisti Biffignandi
Gherard Batisti Biffignandi
KWKuangyi Charles Wei
Kuangyi Charles Wei
JHJoel Hellewell
Joel Hellewell
CJClaire Jenkins
Claire Jenkins
JCJukka Corander
Jukka Corander
KBKate Baker
Kate Baker
Genomic surveillance has transformed our ability to identify and track bacterial lineages and epidemiologically relevant traits. However, surveillance approaches are mainly based on prevalence-based measures, which can obscure the dynamics of traits undergoing rapid expansion within genetically and temporally heterogeneous populations. This limitation is particularly relevant for antimicrobial resistance (AMR), where the emergence and subsequent expansion of newly acquired transmissible traits can generate substantial changes in population-level prevalence. Here, we introduce EMERGENe (https://github.com/gbatbiff/EMERGENe), a phylogenetic framework that combines ancestral-state reconstruction and analysis of phyletic patterns to quantify the emergence dynamics of gained and transmissible traits. Using time-scaled phylogenies and binary presence data, EMERGENe identifies independent phyletic events in which a trait is acquired and subsequently inherited by its descendants, and estimates interpretable Entry and Emergence rates that capture the introduction and expansion of trait-specific populations. We evaluated the performance of EMERGENe using phylogenetic simulations across evolutionary trajectories with different population growth dynamics, and applied our framework to a national genomic surveillance dataset comprising 3,745 Shigella sonnei isolates. Across simulated evolutionary scenarios, EMERGENe was superior to prevalence for discriminating traits undergoing rapid population growth from those with slower or no expansion. Applied to S. sonnei, our method detected previously known epidemiological acquisition of resistance to azithromycin, ciprofloxacin and third-generation cephalosporins, while providing information on their underlying emergence dynamics. Temporal analyses also revealed the progressive expansion of ceftriaxone resistance, overlapping with the increasing replacement of previously highly disseminated azithromycin resistance. EMERGENe also detected emerging and overlooked traits, including a recently described epidemiologically relevant phage plasmid and the qnrS1 gene. EMERGENe provides a complementary approach to genomic surveillance by shifting the focus from static trait prevalence towards the evolutionary processes underlying trait emergence and expansion. Thus, EMERGENe provides a robust quantification method for comparison of trait emergence, and by identifying early signals of rapidly emerging traits, also has the potential to improve longitudinal surveillance, facilitating earlier intervention in the onward transmission of AMR in bacterial populations.

27 March 2026 in BMC Microbiology
Verified
Journal Article
Open access
PDF file
KWKuangyi Charles Wei
Kuangyi Charles Wei
BBBeth Blane
Beth Blane
JTJacqueline Toussaint
Jacqueline Toussaint
SRSandra Reuter
Sandra Reuter
MTMichelle S. Toleman
Michelle S. Toleman
WRWilliam Roberts-Sengier
William Roberts-Sengier
Background: Microbial genome-wide association studies (GWAS) are crucial for linking genetic variation to phenotypic traits in bacteria. However, current tools often involve complex manual processing, limited scalability, and fragmented workflows, which constrain large-scale or routine bacterial GWAS. Results: We developed BaGPipe, an automated and flexible bacterial GWAS pipeline built using Nextflow and incorporating Pyseer for association analysis. BaGPipe integrates pre-processing, statistical analysis, and downstream visualisation into a unified workflow that is reproducible and easy to deploy across diverse computational environments. BaGPipe was validated on a publicly available dataset of Streptococcus pneumoniae whole-genome sequences, and reproduced published findings with improved computational efficiency. BaGPipe was then applied to a dataset of Staphylococcus aureus whole-genome sequences, successfully identifying known and novel antibiotic resistance associations. Conclusions: By offering an accessible, efficient, and reproducible platform, BaGPipe accelerates bacterial GWAS and facilitates deeper exploration into the genetic underpinnings of phenotypic traits. BaGPipe is freely available at https://github.com/sanger-pathogens/BaGPipe .