Winning the Conference Lottery

While talking with graduate students about the upcoming conference for the American Society of Mass Spectrometry, I realized that I could help tell the story of how speakers are chosen for this conference each year.

ASMS is a large assembly, with a yearly attendance in excess of 6500.  That might seem huge, but keep in mind that the recent National Rifle Association meeting in Nashville drew 78,865 people.  Many graduate mentors require their students to submit an abstract to the ASMS meeting as the “price” of attendance.  These abstracts generally lead in one of two directions: A) the abstract is chosen for an oral presentation, or B) the abstract is chosen to be presented as a poster.  Oral presentations are organized into 64 different sessions; ASMS 2015 will be a “multi-track” conference, with eight different sessions taking place at the same time, morning and evening, over four days (8x2x4=64).  Each session includes six oral presentations in 120 minutes (20 minutes for each slot, split between lecture and questions).

At first blush, it might seem like it would be challenging to fill the schedule, since it includes a total of 384 lectures!  A look at the cavernous poster hall, however, tells a very different story.  According to this year’s final program, a total of 2762 posters will be presented (plus a couple allowed to be displayed on multiple days).  If we assume that each poster was also considered for an oral presentation (we know that some are not, at the request of the authors), we see that 7.2 abstracts are pushed to posters for every one that becomes an oral presentation.

I have twice served as a panel organizer for ASMS.  The first time, I played a role in defining the topic for the panel, but the second time I was asked to chair a panel that already existed.  In each case, I remember being faced with around sixty abstracts to fill the six speaker slots for my panel.  A similar number of abstracts had listed my panel topic as a second-best fit.  This is the rough process I used to organize the task:

  1. Glance through the abstracts to sample the gestalt.  My idea of what the panel includes may not match what the submitters thought it should.  When I chaired a panel most recently, the “informatics of protein identification” faced an onslaught of abstracts in proteogenomics (essentially the combination of proteomic and genomic data).  The subject clearly possessed a lot of momentum, so I needed to shift my expectations in that direction.
  2. Triage, triage, triage!  Some researchers do not market their work well, burying the lede rather than explaining what is important right up front.  Some researchers make claims that are downright silly, such as claiming an infinite improvement with their algorithm because the competing software produced zero hits.  Sometimes the results are rather ho-hum, while others are so implausible as to be dismissed outright.  I produce a shortlist from my set of abstracts that includes around half the original number.  Frequently that set is split into categories; I might decide to “spend” two presentations to be drawn from a set of six possibilities on a particular topic.
  3. Balance the speakers.  If your panel includes only senior investigators in your field, you will send a discouraging message to junior researchers who are trying to make a dent.  In my experience, graduate students and post-docs are more likely to present high-risk material.  If you include only males in your speaker roster, you risk winning the Hasselhoff award.  Sometimes a phone call to a colleague can be helpful in assessing the presentation skills of a potential speaker.
  4. Second-guess your friendships.  By the time you are selected to make these decisions, you will have developed a rich network of relationships with investigators in your field.  If we all selected only speakers from labs that we know, however, our fields would become cliques and we would stagnate.  In the past I have intentionally selected someone I did not know well over a long-time friend  just to ensure the audience got the chance to see someone new.

What can you do to increase your chance of selection?

  • Make the selector’s job easy: if I must invest serious time into an abstract to determine why it is important, I will simply assume it is not important.  Borrow strategies from our friends in journalism to put the most significant findings first.  I frequently see scientific writing from junior investigators that assumes prior knowledge of their work; you should explain your work as though the selector has never heard of it.
  • Be concrete: abstracts generally provide enough information to let the selector determine whether an idea is pie-in-the-sky or has substantial data behind it.  If you are working in bioinformatics, you will be expected to show the algorithm enables new biological insight, not just lower run-times.
  • Be visible: when you ask a question at a conference, introduce yourself, and give your affiliation.  Make productive contact with leaders in your field; most investigators are willing to answer well-formed questions that reflect knowledge of our work.  When a high-profile speaker visits your institution, ask to be part of a lunch meeting.
  • Do valuable work: in my field, “YA” is an abbreviation for “yet another,” as in “yet another database search algorithm.”  Novelty is your friend.  Work to resolve a long-standing barrier in biotechnology is generally useful.  Of course, your chance of selection is better if your experiments work out beyond your wildest dreams.

May your efforts lead to amazing science!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s