ASHG Posters: The Agony and The Ecstasy

ASHG is a huge meeting, probably the second largest I've ever attended after ASCO.  ASBMB is similar in size perhaps, though I think a hair smaller and definitely a bigger than ESHG.  And certainly multiple AGBTs in scale.  And that is crashing in on me now.

I'm often wearing multiple hats at genomics meeting, at least figuratively - physically I stick to a single announced hat so people can find me.  At AGBT I'm often alternating between my work mind, my bloggin mind and just general curiosity - as well as catching up personally with people I know.  ESHG was a bit more relaxing last summer as almost nobody knew I was there and I could focus on rare disease talks.  But ASHG is being footed by work and is prime ground for making important contacts, so I will be a Biloban first this time.  And since Ginkgo HQ is just down the street, I'll be bouncing back there for some meetings with conference attendees so we can show off our highly automated RAC (Reconfigurable Automation Cart) laboratory.  No rest for the wicked.

But I still like a good poster.  Poster sessions are good ways to skim a lot of information and engage all my interests  I have - surprise! - strong opinions about good poster design.  I like a dense poster which I can follow solo - just as it is interesting to look up at Michelango's  Sistine Chapel ceiling, but also enjoy getting a guided tour - particularly from junior scientists and trainees who otherwise wouldn't get to engage on their work with greybeards =like me.  Also like the Sistine Chapel - without prep from a guide who knows their art history motifs (silencio in the chapel itself!) few will understand the artistic motifs used.  On the other hand, nobody ever paints scientific posters as frescos, nor is it typical to compose one while lying on one's back! 

There was a trend - I'm hoping we can call it a passing fad - a while back to make very minimalist posters with only high level thoughts - not quite reduced purely to memes and emojis but in danger of sliding that way.  No thank you!

ASHG presents a serious problem.  The only way to get to poster abstracts I've found so far - but haven't really looked - is through the official App.  And that access leaves almost everything to be desired.  There's just a rudimentary keyword search and otherwise you must scroll down through the abstracts.  And scroll. And scroll. I flagged 244 of high interest to me based on the title.  My gut estimate of what fraction I flagged is between a fifth and a tenth, so there must be 1000-2000 total posters

But it gets even more challenging.  Each poster number has a suffix of 'W' 'T', or 'F'  - perhaps my reaction when it hit me what this suggested.  Each poster is up for only one day!  At ESHG each poster author had a designated session to be stationed by their work, but the posters were up the entire time.  Not here.  Looks like I have only 73 to race through on Friday; Wednesday there are 74 to warm up with but Thursday will have a punishing 93. 

I certainly won't have time to walk by all to note any I missed in my scan.  That was painful enough - was doing it during my commute on the train & several times the app reset the list - so I had to scroll madly for a long time to get back to where I stopped.

You'd think in this day and age one would have Chat GPT access to the abstracts - show me all the PacBio abstracts or all the ones with long read topics or so forth.  Just attempting to find all the PacBio took several tries.  Or if one could flag abstracts of interest there could be a "attendees who flagged this abstract tended to flag these too".  Even just some simple text clustering would be a huge improvement over the presented scheme.

Even just being able to partition the abstracts by day is missing.  I did this by copying the poster numbers and then using grep, but that's not in everyone's wheelhouse.

Posters are a valuable way to transmit scientific information - but only being able to either graze them in person or scroll interminably just doesn't do them good service.  I have the same complaints about other meetings; ASHG is finally breaking the camel's back because it is so large.  Better ways to direct meeting attendees to posters of interest would increase the value of these conferences.  And doing that with AI would certainly fit the zeitgeist.

Addendum: After writing this yesterday but reserving time to expand, I ran into a colleague I hadn't seen in decades - he was an undergraduate researcher in my graduate lab.  He had seen all the abstracts as a giant PDF - but I haven't found it.  So somebody could load these into an LLM - if you are adept at this please do so!