First, a bit about the meeting. The meeting is formally called the "Human Microbiome Research Conference" and more information about it can be found here. It is directly tied to the NIH "Human Microbiome Project", also known as the HMP, which has been in operation for a few years now. The HMP is one of a small number of NIH "Roadmap" initiatives (these are also known as NIH Common Fund Projects). These are cross cutting projects that are funded outside of the normal NIH departments/centers. The HMP started a few years ago and is focused on studies of the communities of microbes that live in and on humans. The HMP has so far funded a wide array of projects including some big scale and some smaller scale. This meeting is I think the first to try to cover the diversity of different projects funded by the HMP at once.
Day 1 covered a bit of introduction and then a diversity of HMP related and some not so related (but good) talks. The talks were perhaps a bit too reviewy for my liking, but I think actually, all of them were interesting. Some lessons I got from today include the following:
- The human microbiome is becoming a fascinating area of research with an ever growing set of data to look at
- The data for the HMP seems to be openly available, which is good. Reference genomes can be found here. Some rRNA data can be found here. Strains of microbes are available too. Not sure where the metagenome data.
- Ease of access and use of data is more important than just access to data; the HMP DACC is doing a decent job with helping access data.
- As important as access is metadata about samples. Not sure how much of this is available right now nor how easy the metadata is to make sense out of.
- As with most (all?) large scale, top down projects, there are multiple areas where improvements could be made in communication and engagement with the broader community. It seems like the HMP is working hard on this issue.
- There are some possible complicated issues around release of microbiome data and medical records from people
- There is still a big risk in overselling the potential benefits of microbiome research
- Correlations ≠ causation. Sorry I had to put that here. See #5 above.
- Analyzing and making sense of metagenomic data is still very very hard
- We desperately need more ecology driven studies of the microbiome
- To me, the HMP should really try to mimic the human genome project and focus on producing reference data (genomes, metagenomes, and rRNA) for everyone to use. Trying to do complex scientific/clinical studies in this project seems inappropriate. We all need the baseline to do the science.
- As with every meeting, the best stuff that happens is in between talks.
- Having the meeting room be something like 40 °C is probably not the best idea nor use of resources (hotel issue, not HMP issue).
Those are some of the lessons I am thinking about now, a few hours after the last talk. But if you want to get a "real time" feel for the talks, the best way to do this, if you weren't here, is to look at twitter posts about the meeting. If you do not know, the common practice these days is to use a code within twitter specific for the meeting called a hashtag. For this meeting the hashtag is #HMP2010 and you can find the tweets about the meeting easily by searching twitter for this code. There were other twitter posts about the meeting, but may be somewhat hard to find b/c they did not use this code. Such is life I suppose.
Anyway, if you want to see all the tweets from the first day of the meeting with this hash tag I have appended them at the bottom of this email. This includes anyone who may have reposted (aka retweeted) these tweets to their twitter feed.
As I have tweeted many meetings I guess I am used to various aspects of such activities but many out there clearly are not. Live tweeting a meeting is a rough thing in many ways, at least for me. I want to give people a feel for the meeting, as it happens. I want them to know what I actually think about talks, at least within some reasonable limits. But alas sometimes, hopefully not too often, I get things wrong. And sometimes I post something obnoxious. And sometimes I miss key points. To me, this is analogous to the conversations people have about talks all the time. Overall, I think mostly I do an OK job tweeting meetings. But occasionally I write something that does not sit right with others or myself. And alas, today has one such tweet (well, only one I know of right now).
It happened during Brice Birren's talk. Birren, from the Broad, was discussing a few different things including studies done by the Broad in which they have tried to compare and contrast and use rRNA PCR studies done at different centers associated with the HMP. Much of what he was discussing was technical details of the control experiments they did to assess how variable the results were between centers. I felt at the time that he was placing this work in enough of the context of other rRNA studies of the human microbiome. And I tweeted this feeling.
phylogenomics
Birren presenting clustering of samples based on rRNA analysis - but could do much better referencing prior lit on this #HMP2010
phylogenomics
Birren - given that they trust their PCR, now comparing many human samples - seems like they are way behind the field here #hmp2010But in the end, after discussions at the bar later, I think I may have missed the point of his talk. I thought at the time that he was discussing solely new findings and new analysis tools that they developed. To me, I did not like that he did not spend much time discussing other analysis tools nor how their work compared to other studies of the human microbome. And that gnawed at me.
But in retrospect, I think perhaps he was focusing more specifically on the comparison across the centers. In that context, the way he laid out his talk and what other work he referenced makes more sense. I think perhaps he could have still placed things in a broader context but my comments in retrospect were a bit over the top and unnecessary. As I said above, I think it is useful to try and post what I am actually thinking at the time. I try to filter this if I know what I am thinking is rude, biased, obnoxious, etc; but I do post critiques if they seem relevant. But my filtering was a bit off here. Alas, now Bruce (if he reads twitter) probably wants to dump a beer on my head and people from Broad think I am a putz (well, some may have thought that before). I guess I may have to change the tuning on my filter a bit ... but I still will try and post what I think at the time. It is a fine balance I do not always do precisely ... off to sleep and in the AM - Day 2 plus time to make some in person apologies ...