CBRI/SysGen Seminar – Lachlan Coin – 18th November 2016
Rapid detection of antibiotic resistant bacterial infection
Clinical features do not accurately distinguish bacterial from viral infection. Moreover, detection of antibiotic resistant bacteria typically requires more than 48 hours, largely due to the bacterial culture step. As a result, patients presenting to intensive care units with fever are often given antibiotics inappropriately, while occasionally patients with a bacterial infection are sent home without appropriate treatment.
In this talk I will present the work we have been doing to develop rapid diagnostic tests of bacterial infection and antibiotic resistance. I will first present our work in identifying a host immune-response transcriptional signature which distinguishes bacterial from viral infection in children with more than 90% sensitivity and specificity. I will also describe the novel variable selection algorithm we developed in order to obtain this minimal signature. Secondly, I will describe our work to predict antibiotic resistance directly from biological sample using nanopore sequencing coupled with magnetic nanoparticles for enriching bacterial cells. I will describe the streaming algorithms for strain and species identification, scaffolding, and detection of antibiotic resistance genes from real-time nanopore sequence data.
Lachlan completed a Bachelor of Science at ANU, majoring in Mathematics. After several years out of science working in consulting, he went to the Wellcome Trust Sanger Institute to do a PhD in bioinformatics. Lachlan was a research fellow at the School of Public Health, Imperial College London, largely working on methodology for genome-wide association studies. He returned to Australia in 2012 to start a group at the Institute for Molecular Bioscience, University of Queensland where his group works on developing genomics and bioinformatics tools in infectious disease and cancer.