Search bioRxiv⌕ Search

Biology subjects

Blakeway, L. V.

Publications and source records attributed to Blakeway, L. V..

3 recordsLinked to original sources
bioRxiv2023

Predicting Pseudomonas aeruginosa drug resistance using artificial intelligence and clinical MALDI-TOF mass spectra

Matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) is widely used in clinical microbiology laboratories for bacterial identification but its use for prediction of antimicrobial resistance (AMR) remains limited. Here, we used MALDI-TOF MS with artificial intelligence (AI) approaches to successfully predict AMR in Pseudomonas aeruginosa, a priority pathogen with complex AMR mechanisms. The highest performance was achieved for modern {beta}-lactam/{beta}-lactamase inhibitor drugs, namely ceftazidime/avibactam and ceftolozane/tazobactam, with area under the receiver operating characteristic curve (AUROC) of 0.86 and 0.87, respectively. As part of this work, we developed dynamic binning, a feature engineering technique that effectively reduces the high-dimensional feature set and has wide-ranging applicability to MALDI-TOF MS data. Compared to conventional methods, our approach yielded superior performance in 10 of 11 antimicrobials. Moreover, we showcase the efficacy of transfer learning in enhancing the performance for 7 of 11 antimicrobials. By assessing the contribution of features to the models prediction, we identified proteins that may contribute to AMR mechanisms. Our findings demonstrate the potential of combining AI with MALDI-TOF MS as a rapid AMR diagnostic tool for Pseudomonas aeruginosa.

microbiology↗
bioRxiv2023

Genomic dissection of endemic carbapenem resistance: metallo-beta-lactamase gene dissemination through clonal, plasmid and integron transfer pathways

Infections caused by metallo-beta-lactamase-producing organisms (MBLs) are a global health threat. Our understanding of transmission dynamics and how MBLs establish endemicity remains limited. We analysed two decades of blaIMP-4 evolution in a hospital using sequence data from 270 clinical and environmental isolates (including 169 completed genomes) and identified extreme gene promiscuity across 7 Gram-negative genera, 68 bacterial strains and 7 distinct plasmid types. An initial multi-species outbreak of conserved IncC plasmids (95 genomes across 37 strains) allowed endemicity to be established through the ability of blaIMP-4 to disseminate in successful strain-genetic setting pairs we termed propagators, in particular Serratia marcescens and Enterobacter hormaechei. From this reservoir, blaIMP-4 persisted through diversification of genetic settings that resulted from transfer of blaIMP-4 plasmids between bacterial hosts and of the integron carrying blaIMP-4 between plasmids. Our findings provide a framework for understanding endemicity and spread of MBLs and may have broader applicability to other carbapenemase-producing organisms.

microbiology↗
bioRxiv2022

Characterisation of key genotypic and phenotypic traits of clinical cystic fibrosis Staphylococcus aureus isolates

IntroductionOne third of people with CF in the UK are co-infected by both Staphylococcus aureus and Pseudomonas aeruginosa. Chronic bacterial infection in CF contributes to the gradual destruction of lung tissue, and eventually respiratory failure in this group. Gap StatementThe contribution of S. aureus to cystic fibrosis (CF) lung decline in the presence or absence of P. aeruginosa is unclear. Defining the molecular and phenotypic characteristics of a range of S. aureus clinical isolates will help further understand its pathogenic capabilities. AimOur objective was to use molecular and phenotypic tools to characterise twenty-five clinical S. aureus isolates collected from mono- and coinfection with P. aeruginosa from people with CF at the Royal Victoria Infirmary, Newcastle upon Tyne. MethodologyGenomic DNA was extracted and sequenced. Multilocus sequence typing was used to construct phylogeny from the seven housekeeping genes. A pangenome was calculated using Roary. and cluster of Orthologous groups were assigned using eggNOG-mapper which were used to determine differences within core, accessory, and unique genomes. Characterisation of sequence type, clonal complex, agr and spa types was carried out using PubMLST, eBURST, AgrVATE and spaTyper, respectively. Antibiotic resistance was determined using Kirby Bauer disk diffusion tests. Phenotypic testing of haemolysis was carried out using ovine red blood cell agar plates and mucoid phenotypes visualised using Congo red agar. ResultsClinical strains clustered closely based on agr type, sequence type and clonal complex. COG analysis revealed statistically significant enrichment of COG families between core, accessory and unique pangenome groups. The unique genome was significantly enriched for replication, recombination and repair, and defence mechanisms. The presence of known virulence genes and toxins were high within this group, and unique genes were identified in 11 strains. Strains which were isolated from the same patient all surpassed average nucleotide identity thresholds, however, differed in phenotypic traits. Antimicrobial resistance to macrolides was significantly higher in the coinfection group. ConclusionThere is huge variation in genetic and phenotypic capabilities of S. aureus strains. Further studies on how these may differ in relation to other species in the CF lung may give insight into inter-species interactions. Data summaryThe assembled GenBank (gbk) files for all clinical isolates in this study have been deposited in ENA under the study accession PRJEB56184, accession numbers for each of the twenty-five clinical isolates have been provided in Table S1. The reference strains were collected from the NCBI BioSample database (www.ncbi.nlm.nih.gov/biosample): MRSA_252 (NC_002952.2), HO 5096 0412 (NC_017763.1), ST398 (NC_017333.1) and NCTC8325 (NC_007795.1).

microbiology↗