About me

Since 1998 I have been a regular user of macromolecular crystallography (MX) beamlines at more than five synchrotrons worldwide. I have spent over 10,000 hours collecting data and deposited more than 60 structures in the Protein Data Bank. My experience covers the whole pipeline: crystallisation, crystal harvesting, data collection and reduction, structure solution, refinement and analysis. I have co-authored more than 35 peer-reviewed papers, cited over 1,500 times since 2003.

Highlights

• Contributing to membrane-protein work on G-protein-coupled receptors (GPCRs), research recognised by the 2012 Nobel Prize in Chemistry awarded to Brian Kobilka.
• Shedding light on electron transfer between cytochrome c and respiratory complex IV, one of the key energy-conversion steps in cellular life.
• Studying a membrane kinase in remarkable detail using both synchrotron and X-ray free-electron laser (XFEL) radiation.
• Showing that small membrane peptides can be solved by crystallography using the lipidic cubic phase.

In 2011 I moved from an academic post in structural biology (Marie Curie Intra-European Fellow at Trinity College Dublin) to an instrument scientist position on the MX beamlines at the Australian Synchrotron.

Within six months I had become the go-to troubleshooter for hardware and software problems. Attention to detail, thorough documentation of both failures and procedures, and a strong commitment to quality control (QC) ran through all my work. I was closely involved in:

• the software and design of a microservices-based beamline control system to replace Blu-Ice;
• hands-on upgrades of optical elements (a double-crystal monochromator for the microfocus beamline) and sample changers (SAM robots);
• project-managing a new goniometer, from design and manufacture through installation and acceptance testing;
• the detector upgrade to an EIGER 16M, described in two publications.

The role also sharpened skills such as time management, teamwork, creative thinking and conflict resolution.

Since 2019 I have been a Senior Beamline Scientist at Diamond Light Source, the UK’s national synchrotron, in Oxfordshire.

I started on SGI’s IRIX flavour of UNIX and quickly moved on to installing and configuring Linux. My first paid work was setting up SUSE Linux computers for undergraduate chemical engineering classes; later I set up and ran Red Hat, Debian and Slackware LAMP servers. I am a keen supporter of free and open-source software and have been a member of the Free Software Foundation since 2003.

Since 2003 I have run three virtual private servers hosting more than 60 domain names. This has given me broad experience of cPanel/WHM hosting, troubleshooting operating systems, and investigating and closing security breaches across Apache web servers, Exim mail servers, DNS, firewalls and more.

I hold the Red Hat Certified System Administrator (RHCSA, 2012) and Red Hat Certified Engineer (RHCE, 2013) certifications.

From 2012 to 2019 I was the de facto system administrator for the two MX beamlines at the Australian Synchrotron, where I gained extensive experience with virtual machines and Linux containers on the open-source Proxmox cluster. Between 2014 and 2018 I led the move of all the beamline computers from physical machines to a fully virtual infrastructure.

I started with Microsoft Visual Basic, building interactive applications on top of Microsoft Access databases. Running LAMP servers then led me to write dynamic, database-driven PHP/MySQL websites, and from 2004 to 2008 I was webmaster for several sites built on Joomla, WordPress and phpBB.

In 2013 Andy Bulka introduced me to Python with a six-week course, one full day a week. Two years later a second six-week course covered advanced Python and object-oriented programming. Those courses, years of hands-on work on the Australian Synchrotron MX beamlines, and what I learned from the brilliant coder Nathan Mudie changed the way I see things. I was no longer just a user of software: I could use it to drive hardware, collect better diffraction data and improve data analysis.

Selected projects

• An attenuator wheel to control X-ray beam transmission at different energies (Beer–Lambert law, with hardware control through Python and EPICS).
• Moving the auto-processing software to Docker containers on Kubernetes.
• Rewriting the Python data-collection server for the move from CCD-based ADSC detectors to an EIGER pixel-array detector.
• Extending a Python beamline library that keeps applications independent of the instrument they run on.
• Designing the gradual replacement of the monolithic Blu-Ice with a microservices-based ecosystem.
• Developing and maintaining BeamlineSetup, a QC-driven beamline setup GUI that delivers high-quality calibration to users. Its successor, MX Beamline Setup, is now developed at Diamond Light Source.
• Implementing diffraction rastering for sample centring on the MX beamlines.