Stuff that occurs to me

All of my 'how to' posts are tagged here. The most popular posts are about blocking and private accounts on Twitter, also the science communication jobs list. None of the science or medical information I might post to this blog should be taken as medical advice (I'm not medically trained).

Think of this blog as a sort of nursery for my half-baked ideas hence 'stuff that occurs to me'.

Contact: @JoBrodie Email: jo DOT brodie AT gmx DOT com

Science in London: The 2016 scientific society talks in London blog post

Sunday, 31 October 2010

Nobel winner Barry Marshall and the unfriendly bacteria

This lecture took place in September 2010 and is now available online. I started typing up my notes soon after but have only got about half way through. Normally that would mean it would be ages before I finished and posted them as a blog post, but with my new 'let's serialise everything' plan I'm posting unfinished things and coming back to them later. Not sure when I'll finish it though, as I may have carelessly lost the remaining notes pages, oops.

Barry Marshall
'Past lessons and future opportunities for Helicobacter pylori'
UCL Prize Lecture, 20 September 2010

As soon as we started it was clear we were in for good-humoured fun - the audience was in a cheerful mood and I think it's safe to say we were a 'good crowd'. In number too, there was barely a spare seat in the lecture theatre. As I work at UCL I'm automatically on a list to receive information about our events - if they're public then I'll usually share them on my Posterous blog, but there's a listing of public events here. The event took place in one of the lecture theatres in the gorgeous Cruciform building, designed by the same chap (Alfred Waterhouse) who designed the Natural History Museum.

To start we had a brief introduction from someone who ran through Barry's biography, wryly admitting he'd got it from Google and Wikipedia, which got an appreciative chortle. We also saw a list of previous UCL medal winners, including a number I'd heard of.

As the introducer was about to hand over to Barry he tried to set up the presentation for him, struggling gamely with an animation of the 3D structure of the protein urease that was part of the powerpoint, and expressing amused irritation that the software wasn't yielding to his commands. Barry's 'I can do that', from his audience seat, got a welcoming giggle and he took to the stage.

Barry jokingly suggested that if you're going to make a Nobel-worthy discovery do it while you're young as it'll probably be a couple of decades before you'll get the prize. He and the team he was working with published their findings on the link between H. pylori and gastric ulcer in 1984, he won the Lasker Award in 1995 and the Nobel Prize in 2005. He showed us a rather nice cartoon slide of a couple of the bacteria - this was developed with a graphic designer and submitted to a competition at The Scientist (it's a finalist).

Around the time of the discovery there were a number of possible options for 'things that cause gastric ulcers', stress being thought to play a major role. No-one was terribly excited by the prospect of another possible cause, so the discovery seems to have arrived against a backdrop of 'meh'. Barry then put up a quote from Daniel Boorstin who said that 'the obstacle to discovery is the illusion of knowledge' - there was no particular imperative to search for other causes.

The impact on individual and public health has been notable, not to mention on lost productivity. He suggested the scenario of busy stressed executives with an ulcer being ordered by their doctors to retire early and take to a life of beach-ly leisure. More seriously, ulcers had a major cost impact and wrecked lives with several thousand deaths every year in the 1980s.

While there wasn't much impetus to find alternative causes for gastric / peptic ulcers bacteria were already low on the list of possibilities. The medical textbooks of the time held that the stomach, being highly acidic, was sterile and any bacteria introduced would be killed off. One of the reasons for acidity is to extinguish some newly met bacteria to 'sterilise' food and prevent gastroenteritis, but as we all know now the stomach contains many different bacteria which are considered essential for health. Extremophiles have also been discovered which include bacteria that can survive and thrive in conditions previously considered as unable to support life. Most bacteria are killed at pH4 and the stomach's acidity is at pH1.5 so it's not too surprising that bacterial infection wasn't considered as a likely candidate.

Barry showed us a slide of biopsies taken from 100 patients (quipping that they used 100 patients as they didn't have a computer at the time and so it made the maths easier) where 13 out of 13 patients with peptic ulcer had these twirly bacteria in their biopsies. I couldn't help but be reminded just a tiny bit of a certain struck-off someone who looked at the guts of children and drew some wrong conclusions.

He and a colleague tried to get this data into a meeting but were rejected - he showed us the rejection letter, and recommended that people kept theirs to amuse crowds in future. Apparently of 67 abstracts submitted the organisers could accept only 56 ;)

I missed an interesting aside into statistics, so forgive if this is a little bit wrong, but in the list of 100 patients they spotted that almost everyone with 'oesophagus abnormal' which is also known as acid reflux didn't have H pylori - the level of significance suggested thateliminating H pylori might increase the risk of acid reflux (but I might be wrong).

The next article he showed was 'An attempt to fulfil Koch's postulates for Campylobacter pyloridis' from Medical Journal of Australia, 1985. Koch came up with a series of tests, in the 1800s, in order to demonstrate that a condition had a bacterial cause. First catch your bacteria (from someone who has the disease you're studying), culture it, infect someone or something, wait for your disease to appear then collect evidence of bacterial infection.

This was the next step for H. pylori and Barry drank some (it turned out to be two Petri dishes' worth, someone asked at the end!) to show that they could survive and colonise in his stomach. He showed a biopsy photo which clearly illustrated an infection, and some pretty annoyed stomach lining cells. The infection was accompanied by nausea, vomiting and other gastric disturbances - I think he was fairly prompt with a course of antibiotics after though, so probably didn't have a full blown ulcer. He said he was surprised by how ill he felt - many people have asymptomatic H. pylori for (probably many) years before ulceration shows up.

His suggested explanation is that people acquire their infection in childhood, have a few days of being sick but then recover though the bacteria remain, ready to create ulcers at some future date.

It took quite a while for anyone to be particularly impressed by this and I expect that somewhere there's a seller of nonsense alternative therapies who mentions Barry Marshall in the same breath as Galileo as someone who was ultimately proved right.

In 1997 Abbott, who marketed a treatment for H. p took matters into their own hands and commissioned a cartoonist to visit Marshall's lab and create a storyboard. There was a fantastically lurid drawing of Barry holding a beaker of bubbling green liquid about to drink it saying "there's no other way" with a colleague looking on saying "you're crazy" (17m10s). The resulting cartoonified story was sent to a 150,000 doctors and was apparently very successful in convincing them that peptic ulcers should be treated with a course of antibiotics. Obviously I wondered about other uses of cartoons in communicating science, either in medical education or public engagement!

Apparently the bacterial solution was actually brown but the cartoon people had some involvement in "The Incredible Hulk" and went instead with green...



My l'esprit d'escalier question was "Why did people who had to take antibiotics for other things not notice their ulcer getting better?"

Saturday, 30 October 2010

Episode III of #IASBB evening event - beyond science blogging

I'm lucky enough to get to go to all sorts of interesting events at which I make lots of notes. The 'writing them up' bit doesn't seem to go quite as well so I've decided to work in harmony with my inability at this, and serialise my posts. Here's the third in the 'series' of posts about the evening event a couple of weeks ago (oops!) after the #IASBB, I'm a Scientist Beyond Blogging event.

Previous #IASBB posts were
At the evening presentations we heard back from the different groups who'd spent the afternoon looking at ways in which scientists might use the online world even more to engage with people, including being more politically involved, ie taking things 'beyond blogging'.

Alice Bell reported back for her group - they'd been looking at science and politics and how the two communities might work together. I think the British Science Association have or had a short-term placement scheme for MPs and working scientists but I don't know how many people could feasibly be accommodated on the programme.

Apparently the group came to the conclusion that we should all watch the West Wing - I expect there'd be uniform agreement with this as it's wonderful, I have the box set. So many great instances of affection for science and knowledge - possibly the episode that makes me swoon the most is 'Galileo' - rest assured, CJ "says it right" at the end.

The group also came up with the idea for some sort of online "Campaigning for scientists, for dummies" where scientists can pose questions, answerable by others. The Science is Vital campaign (and there'll be a future blog post at some point on the fab presentation by Jenny Rohn, Richard Grant and Shane McCracken) involved a steep learning curve for the organisers in terms of arranging the rally outside the Treasury. Lots of community input.

I think I missed the comment at the time but possibly it was Alok Jha who tweeted along the lines that science bloggers should get their posts onto places like Conservative Home and target other blogs like that, reaching a different crowd etc.

Finally, the group had discussed the notion of a science-specific campaign roadshow telling people about science and citizenship, and policy. I think there may have been ideas for a travelling bus :)

Sunday, 24 October 2010

Abandoned Britain - half day nerd trips

Shortened link for this post is http://is.gd/gi6Gu

EDIT: 13 April 2011
Nerdytrips - the making of nerdmap (the map expands what's in this post to include the rest of the UK, world... etc.) describes what happens after Ben Goldacre posted "Nerdy Day Trips - tell us about yours, we'll build an archive #nerdytrips" :)

--------------- Original post --------------

A few weeks ago, Ben G asked Twitter followers for suggestions for a weekend half-day out. The suggestions were pretty interesting, and I have a file full of collected tweets from which I'm excerpting the relevant bits and pieces, as promised here.

It's occurred to me that this post is also a showcase for Google's capacity to search historic tweets. To get the URL for the original suggestion I've used Google's updates search facility (which currently goes back as far as February 2010 but is planned to stretch to the first tweet in March 2006). Using this method I've also picked up stuff that I didn't manage to collect the first time. For that reason I wish I'd used a different name as some of these suggestions are great but don't really fit within the 'abandoned britain' headline, oh well ;)

Sites that will signpost to other suggestions

Series 4, episode 4 (Dover to Isle of Wight) of BBC's coast features a piece by Neil Oliver travelling to Dungeness on the Romney, Hythe and Dymchurch railway, and Alice Roberts at Highdown/ The Needles looking at rocket testing.

On a very small scale of mild abandonment, I love the Terma radar (for the aircraft at City Airport I believe) to be found in a lovely but lonely part of Greenwich Peninsula

All the 'suggested by' links will take you to the person's original tweet - often worth a look as there's a bit more info there. Some of the suggestions may not have been entirely serious though... I'm also not sure about Lee / Lea although there's a florists in Lee (the one that's near Blackheath) which is called Fleurs de Lee :)

Miscellaneous (ie stuff I can't quite pinpoint)

  • "undeveloped area, north of the Thames, east of Tower Bridge" suggested by @anthropith
  • "deep level shelters in south London" suggested by @nathanbroon
  • "Have you thought about a trip on the Bakerloo line?" suggested by @mhoulden
Non-UK
  • "best railway graveyard ever is near the Salar de Uyuni in Bolivia" suggested by @Jesssal
Previous edits
EDIT: 7 January 2011 - This morning @imaginarygf tweeted the following, which is a list of quirky museums in London, some of which I've been to. Don't think it mentioned the Hunterian in London but I am dashing out of the door....

"Oooooooo RT @aleksk: o boy o boy o boy... RT @LDN: London's quirkiest museums http://LDN.in/FPKovS (via @visitlondonweb) (hey! @bengoldacre)"

Saturday, 23 October 2010

"Bollocks Watch" - feedback to evening event at #IASBB on proactive science stories

Time for another teeny blogpost on #IASBB. I think serialisation of my blog posts is likely to work better in terms of me actually getting stuff down ;)

This follows on from my post yesterday although the discussion mentioned below actually preceded the discussion mentioned in yesterday's post. At the evening reception for the 'Beyond Blogging' event hosted at the Wellcome Trust by Gallomanor / I'm a Scientist we heard feedback from discussions at the afternoon sessions, which were looking at ways to engage people slightly more actively with science (and scientists ... with public) via online activities.

The first person to report back was Paul whose group had come up with 'Bollocks Watch', a sort of 'proactive version' of some of the sites already in place that react to bad science, such as Behind the Headlines. No slight towards BtH is intended by my use of the probably a bit loaded terms 'reactive' and 'proactive'. I'm a huge fan of BtH and use it frequently at work - even went to the celebratory party a few weeks ago.

I think I'd like to hear more about this Bollocks Watch - the more I write up these mini posts the more I wish I'd been at the earlier part of the event.

My first reaction was - is this really that new? Everyone seems to be addressing the terrible reporting of science in the news™ (I'm not sure if we've stopped yet to notice if it's got better, or if we just notice the bad stuff without remembering the good).

Working for Diabetes UK I do tend to notice any bad stories about diabetes, and health stories - it's my job to find out about what's going on and what people will contact us about. Well-reported clear stories probably don't generate as many calls or emails, or a response that requires a sort of Behind the Headlines explanation. I've also written on this blog (somewhere towards the back) that sometimes the article is fine and people just misunderstand something in it. I wonder what the situation is for science stories that have nothing to do with health.

Anyway - I'm not pretending by any means that all is perfect in science communication via the newspapers etc. And I'm pretty sure we're going to put on our website a guide to spotting terrible science reporting in the news fairly soon ;-)

Paul's group's aim was to counter misinformation arising from bad science reporting (am vastly in favour of correcting misinformation obviously) but more by being creative - generating a story and getting it out there, in the hope of replacing the bad story with a good one. This would involve having a recognised 'brand' and creating an active message of positive stories to 'make the truth better than the fiction'. Paul talked a little about the viral strategies used by 4chan and popbitch (I'm not familiar with these sites to be honest) and having a site where the public can rate reporters, publications and their content (yikes!).

As always I wondered how this might fit in with ventures already in place to address misinformation - I don't know of any that are actively retaliating by trying to get a 'more correct' version into the press, or if that would work. Presumably the initial press release (assuming it's not a fiction itself) is going to be telling a reasonable story (I know, I know) - I'm just thinking aloud how a bad story would be reversed by a second press release.

Are journalists / newspapers / BBC embarrassed by being cited in Behind the Headlines? Does it lead to the story being amended?

On the topic of the reporting of statistical information in the news I'll be interested to see what happens during @scijourntrain's analysis of training opportunities in stats for non-science journalists, and other resources - I added a comment to his blogpost here which I think is relevant.

Another reason why I hate DOI

EDIT 3 March 2011 - Ben Goldacre's just tweeted a link to CrossRef which seems to go one step further than resolving DOIs to a clickable link and resolves a citation to a clickable link. Useful.
------

A friend on Twitter wondered if anyone could get a particular article for them, and it seems I might. If only I could find it quickly.

Clicking on the link given takes me to this page
http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2133.2010.09954.x/abstract

(Edit 3 March - the page now gives details of which journal and issue the article's published in).

but I need to log in using my account, which means that I'm working in a different tab with a different URL across the top and will have to drill into the different volumes to find it. Pasting that URL in doesn't work (tried it) as the fact that I'm logged in doesn't seem to transfer to that page even when opened within the tab that I'd been previously logged into. So drilling it is...

The information I need, now that I'm logged in to the journal's site, is which copy of the journal I'll find it in such as year, month and in some cases the volume and issue number.

The information given tells me that the article was first published online in July 2010, but that doesn't necessarily mean it's in the July issue.

I have absolutely no idea how to extract any useful information from the bit where it tells you how to cite this:
Tsang, M. and Guy, R. , Effect of Aqueous Cream BP on human stratum corneum in vivo. British Journal of Dermatology, no. doi: 10.1111/j.1365-2133.2010.09954.x
I have never yet found an article using its doi.

There are other ways of finding the article - I can 'search within this journal' for the article's title or authors but I'm afraid that the purpose of the DOI system eludes me. It seems to present users with an unpleasantly alphanumeric string to type in (if reading something in print and wanting to investigate electronic versions) and doing so in Google doesn't (yet) take you straight to the article (why? WHY?), but if you paste it into a DOI resolver all will be well.

As a touch typist I can handle typing in pretty much anything, but you do have to attend a great deal more to something that complex, compared to a string of words.

I quite enjoy a search challenge and I'm 98% confident that I'll find it within minutes - obviously I'll have to stop bleating on blog posts first - but DOI SCHMEOI...

To be fair it may not yet have been published in the print version, making my bleat a little less relevant, although from looking on the site usually online first articles are print-published within two months.

Anyway... yes, I've found a PDF copy. Mutter mutter grumble.

Beyond blogging - science engagement online #IASBB

When I started writing this blog post a couple of days ago (procrastination!) I had just returned from a really inspiring evening at the Wellcome Trust's Gibbs building thanks to @imascientist who kindly put my name on the list for the evening roundup. Although I'd love to have gone along for the afternoon too as it sounds really interesting, but I'm more of an evening event kind of person I suppose. This was the evening reception for the 'Beyond Blogging' event.

While there we heard from a number of people - firstly there was feedback from each of the groups from the afternoon session (which I didn't attend), they had each discussed an aspect of science engagement online and were reporting back. I've made rather a lot of notes, and rather than completely fail to type them up all in one go (and then never get around to finishing which is my current blogging tactic) I thought I'd write it up in bits and pieces.

Then we heard from Jenny Rohn, Richard Grant and Shane McCracken with a fascinating day-by-day account of how the #scienceisvital campaign developed from a blog, some volunteers and momentum from Facebook, then Twitter and then - importantly - everyone else via email and word of mouth.

Finally we heard from one of the scientists and one of the teachers who'd been participants in I'm a Scientist, Get Me Out of Here! a while back and Sophia summed up with some pithy thoughts on online engagement.

The bit I wanted to talk about first was one of the five updates from the afternoon sessions. I think I should impress again upon anyone reading that I wasn't at the afternoon discussion and this is my impression of a brief reporting-back - so I might miss stuff.

I think the brief was that people would discuss several ways in which different communities / publics can engage / be engaged with science and scientists while taking things a little further than blogs, ie a bit more involvement and interactivity.

One example that intrigued me was from Jonathan (sorry, don't know his surname) whose group came up with the idea for some sort of database in which would be placed information about funded research in the UK. This would be open to the public, and people could monitor the progress of research and see where their money was going.

There are similar things in place for clinical trials. Trials can be given a unique registration number, which is used in published articles, and you can track its progress. I'm still always pleasantly surprised to see an NCT number in an abstract or article.

About a year ago I collected together 7 databases, mostly for clinical trials (medical research in people) and packaged them together into one URL via the Krunchd service.


Research databases and trials registers
- A non-comprehensive list of research databases, trials databases and other relevant websites.
rdb, research, database, trial, trials, clinical trials, register, registry, registries, registers, clinical, clinical research,
I think a database of other basic research would be rather useful. My question at the allotted time was 'who would input the data?'. I wonder if this might be a bit of a sticking point. Funding bodies, eg charities, have to provide some basic information about where they money's going - there's the requirement for an annual report, but most charities take this a bit further and take the opportunity to talk in more depth about the research they're funding.

This information will include the institution where the work is taking place, the names of the researchers, the title of the project, how long it will last and the amount of money put towards it. That's probably what would be needed (as a minimum) in any other database, but who would input that? The funders, the researchers, volunteers scraping data from these reports, or by accessing institutions' research grants databases (RDBs)?

Many charities, and I'm assuming other funding bodies, will use bespoke RDBs and it would probably help if the contents of each record could be exported in more or less the same way... presumably people wouldn't really want to enter the data into one database and then do the same again for the 'community database'.

Hopefully these are minor issues as I think the idea's a really interesting one.

Here are the words and phrases I wrote down while Jonathan was speaking:
lack of a database for projects, academics, open data community, wider civil society, document progress of a research project, link social media, deposit data sets, API for programmes to interface, 1,000 flowers bloom :)

The database sounds like the sort of thing that gets thrashed out over one of those hack weekends - fingers crossed it does.