Talk, DTU Microbes Initative Conference 2022, Kongens Lyngby, Denmark
Presentation on the topic of my PhD: how to use publicly available sequencing data for genomic surveillance of AMR.
Poster, ASM Microbe 2022 & ISME18, Washington DC, USA
The increasing amount of next-generation sequencing (NGS) data available in public repositories has the potential to provide new ways of monitoring the emergence, evolution, and spread of antimicrobial resistance genes (ARGs) with a large-scale metagenomic analysis. We have downloaded more than 442 Tbp of sequencing reads from 214,095 host-derived and environmental metagenomes from the European Nucleotide Archive (ENA). After quality-checking and trimming the reads, we aligned them against ARG templates from the ResFinder database. Here, we present our network analysis of the co-occurrence of ARGs across a highly diverse collection of samples to find the evidence of ARGs conferring resistance to different antibiotic classes being co-selected at a global scale. Due to the compositional nature and sparsity of working with counts of aligned read fragments, we applied SparCC on counts to infer pairwise linear correlations. The poster presents the preliminary results to demonstrate the feasibility of a large-scale analysis to investigate the occurrence of cross-resistances in different microbial settings.
Poster, 6th Annual Danish Bioinformatic Conference, Aalborg, Denmark
The advancement of antimicrobial resistance genes (ARGs) is one of the biggest threats to public health. Since ARGs can be found in a diverse range of genetic contexts in different species and environments, determining structural elements in the regions up- and downstream, or flanks, of ARGs can provide useful information on their evolutionary history. We will present our analysis of flanks surrounding members of the mcr gene family in metagenomic samples. The mcr genes confer resistance to colistin, a last-resort antibiotic used against multidrug-resistant bacteria. We have identified and assembled 869 mcr positive metagenomes from various origins and locations and have extracted flanks of 1939 different mcr contigs from the assemblies. Plasmid replicons and mobile genetic elements (MGEs) were identified in the flanks. By setting a minimum flank size of 1,000 bp and clustering on the pairwise k-mer distance, we can observe several unique genetic signatures for each mcr gene variant. For example, we can see that only the ISApl1 MGE appears around mcr-1 contigs and IS903 in mcr-9 flanks. As there were only 138 contigs that passed the size criteria, we are in the process of including mcr flanks from 3327 genomes of single isolates retrieved from NCBI to better understand the elements involved in the dissemination of mcr genes. Our presentation will include these results as well.
Talk, Novo Nordisk Foundation Symposium 2021 on Local Landscape and Outlook Regarding Solutions to Antimicrobial Resistance Challenges, Hellerup, Denmark
Talk, VEO Symposium 2021, Online
Talk, ASM NGS 2020, Online
Poster, Antimicrobial Resistance - Genomes, Big Data and Emerging Technologies, Online
The growing amount of NGS data available in public data repositories presents a unique opportunity to explore new ways of monitoring antimicrobial resistance (AMR) across different sample types. The diversity of sample origins, locations and collection dates can provide a detailed picture of the emergence and spread of different AMR genes worldwide. We have downloaded 122 Tbp sequence reads from 116,758 host and environmental metagenomes from the European Nucleotide Archive (ENA), have quality checked and trimmed the raw sequencing reads and mapped them against two reference sequence databases: the Silva database to describe the bacterial content and ResFinder to detect AMR genes. Using compositional data analysis on the output, we have compared the results across different experimental procedures.
We will showcase the feasibility of large-scale metagenomic AMR analysis by focusing on the mcr gene family, which confers resistance to colistin. The spread of mcr genes in recent years poses a significant threat to public health, as colistin is a last-resort against multidrug-resistant infections. By using our data, we have characterized the advancement of mcr genes across animal and human populations. In human samples, mcr-9 were more present as compared to other hosts such as livestock displayed a more diverse distribution of all the genes (p < 0.05). We also see that the mcr genes did not have a global distribution but instead were specific to each demographic region with mcr-9 being more prevalent in North American samples and mcr-1 in Asian samples (p < 0.05). Storing and analyzing large quantities of metagenomic data might not be possible for all research groups, which is why we are planning to setup a database and share it together with the results in a webservice. This will allow the scientific community to explore and use the data in new ways. Furthermore, as new metagenomic data becomes available online, we will add these samples to the database and analyze them.