Chapter_5_LCMS

Generate peakml files with retention time-aligned mass spectrometry peaks.

Raw LCMS data was copied from the LCMS machine onto a memory stick, and converted in MZML format with MSConvert from the proteowizard software suite. This generated 65 MZML files, with file names follwoing the format host_bgc_replicate_runcode. Files were collected into one directory per strain (with each directory holding replicate samples for that strain) - directories were m1146_25, m1152_10, m1152_21, m145_10, m145_15, m145_21, m145_25, tk24_10, tk24_15, tk24_21, tk24_25. RT correction and PeakML file creation were then achieved via pre_process.R, which added directories containing individual PeakMl files, RT-corrected PeakMl files (for each replicate sample), and combined RT-coorected PeakMl files (for each strain) to each of the strain directories described above.

Note - there were 4 hosts (M145, M1146, M1152 and TK24) and 5 BGCs (10, 15, 21, 25, 47). Four cultures did not grow, and so the associated strains were not further characterised (M1146::BGC#10, M1146::BGC#15, M1152::BGC#25, M1152::BGC#15). Methanol extraction was only performed for two of the triplicate cultures of M1146::21, as one was contaminated. All BGC 47 strains were cultured in duplicate. All analyses performed on duplicate files (M1146::21, all BGC 47 strains) yielded an error relating to file writing - I suspect this may reflect RT-alignment issues associated with aligning two files rather than 3. This left the 11 strains above.

Annotation peak picking

This folder describes peak annotation with IPApy2 and subsequent search workflow for BGC-specific, confidently annotated peaks

Untargetted peak picking

This folder describes the search workflow for BGC-specific, confidently annotated peaks