This is my second rotation at the Sanger institute, under the Parasites and Microbes programme. I was supervised by Ste Bentley and Vicky Carr.

Metagenomics is a powerful technique in characterising the microbes present in a sample (like a nasal swab). However, analysing metagenomes is tricky because some microbes, like Streptococcus pneumoniae, are easily mistaken for others. I worked on a protocol to evaluate how accurate a tool is at detecting a specific pathogen. The protocol involves both simulated data (which we can completely control) and experimental data (which we can't really control, but it reflects reality)

Software

Check out the Nextflow pipelines I wrote for this rotation

• metagenome_sim-nf: simulates a metagenome given a pool of fasta genome references.
• bowtie2-nf: a pipeline that wraps bowtie2 alignment and read count summary.
• kraken2-nf: a pipeline that wraps the kraken2 taxonomic classification software.

Acknowledgements

Special thanks to Gerry Tonkin-Hill for the statistical method to simulate metagenomes. Thanks to Luis Gonzalez, Theresa Ochoa, Pablo Tsukayama and Hanson Blake for providing nasopharyngeal swab metagenomic samples.