I think the best way for me to characterise the field of biology in recent years is that it has become digital. In the same way that CCD technology and computational power have become an integral part of doing astronomy over the past two decades, biotechnology is undergoing a revolution brought on by high-throughput instrumentation now generating data on a scale that undoubtedly vastly exceeds the largest ongoing astronomical surveys. With this new paradigm biotechnology is attracting people from diverse backgrounds outside of traditional biology, and this includes physicists and astronomers, where the experience of generating and analyzing ever-growing volumes of data is commonplace.

To date I know of at least five former colleagues who have entered this field in the past two years in various capacities: two in bio-computation, one research scientist focused on instrumentation, one in database design and management, and two (including myself) as algorithm developers focused primarily on raw data analysis. Experience ranges from recent astronomy Ph.D.'s, post-docs and (one) tenured professor.

I think many of us who have pursued astronomy/astrophysics and wish to move onto another career have a concern their background might be perceived as too esoteric or impractical for use in industry. And certainly when I was looking for opportunities outside of astronomy my primary goal was to find a career where my training was of real value and not to simply throw away all that experience. It quickly became apparent that biotech was an exciting field that offered a great deal of potential to fulfill this desire. I am now working at a company (Applied Biosystems) that manufactures a wide range of biotech instruments and systems from mass spectrographs for protein analysis to the gene sequencing machines that have been the workhorse of the public and private Human Genome projects. Probably the most satisfying aspect of the job for me has been finding no shortage of times or opportunities where my experience as an astronomer has had direct relevance.

My career in astronomy was largely observational and in recent years my interests were in microlensing. My first post-doc was at Columbia University working on microlensing in M31, followed by another at the University of Washington working as a member of the MACHO collaboration. Generally speaking, both as a post-doc and graduate student outside of observing, most of my time was spent on image and signal processing of data which included a significant amount of my own algorithm development. This combined with observing experience that exposed me to a broad array of instrumentation and gave me a critical eye towards sources of noise and systematic errors in measurement has been invaluable in my present job.

Clearly general programming experience acquired in almost any basic science graduate education is relevant to the majority of hi-tech jobs. Proficiency in a modern language (C++, java) is an excellent asset (exclusive Fortran programming is not ideal). This can also include scripting languages like Perl which is extensively used in bioinformatics. Here, however are some of the problems I have worked on in the past 18 months where a unique astronomical background has been very relevant:

  • Developing algorithms to extract spectra from 2D LC/MS protein data using similar techniques used in conventional spectroscopy.
  • Feature detection and quantification algorithms in dense microarray images. Problems here involve issues very familiar astronomers working in dense stellar fields: blended features, background subtraction, instrumental calibration, pattern recognition. Anyone familiar with DAOPHOT or DoPHOT has excellent experience for this kind of problem.
  • Simulations of imaging systems to assess properties such as the influence of diffraction and microscope NA on resolution, spatial cross-talk and quantitative accuracy. Again familiarity gained from using tools in packages like IRAF, IDL and MATLAB have made approaching problems like this straight-forward.