Monthly Archives: May 2007

Bioinformatician interactions in bioinformatics support

There was a call from Bioinformatics Zen a couple of weeks ago for a kind of ‘special edition’ for Bio::Blogs. The call requested:

“What would be your best piece of advice you would offer someone starting out in bioinformatics? Databases, coding, organisation?”

As I do bioinformatics support, I thought about some of the things I’ve learned, especially with regards to talking to biologists. So what follows is a non-technical look at how to keep your interactions with your biologists running smoothly.

Bioinformatician interactions in bioinformatics support

When, as bioinformaticians, we think about interactions, it’s probably related to an organisms interactome or what proteins do when they get up close and personal. However the social interactions that we have with our colleagues are what really defines the usefulness of the discipline.

Bioinformatics is a great uniter, bringing together maths, computer science, biology and even philosophy. We are long term ‘interdisciplinarians’ in an age where interdisciplinary science is still almost a buzzword.

Unavoidably, a bioinformatician talks to biologists. Whatever your background it is the quality of these interactions that are the basis for success in your chosen career.

How can you make the most of these interactions?

Know your technology

This might sound obvious, after all this is your speciality right? However there is always the temptation to recommend the latest and greatest to your collaborators. Ask yourself some questions first. Have you actually used it yet? Does it deliver what you expect? Are you chasing technology solutions for technologies sake? Can you explain it clearly to the biologists and outline why you have made your choice?

Know your biology

Even more fundamental one might assume. This is harder if you’re dealing with people from all kinds of lab backgrounds. One day you’re talking to a microbiologist, the next you’re talking to a clinician. It is essential that you have some appreciation of their branch of science. Reviews are your friends, but requesting relevant papers from the biologists in advance will save you hours of grubbing around in Medline. A cursory glance at the labs website for recent publications will ensure that you can hit the ground running. You’re never going to acquire the knowledge they have in a short space of time, understand the limitations of this, and be prepared to leverage their expertise.

Ensure the scientist understands your background

This is incredibly important. To the biologist, you’re a bioinformatician. All too frequently they will make an assumption about your background. You might have a biology background yourself, but they will assume you’re a computer scientist and talk to you like they were lecturing a high school student. Or you are a computer scientist, and the biologist will assume that you’re intimately familiar with the regulation of pathway X in organism Y. Make your experience in the field clear, this helps manage expectations and avoid false starts.

Manage expectations

A lot of biologists don’t appreciate the back-end complexity of bioinformatics work. The applications they’re used to are simply black-box solutions to them. Often they think that bioinformaticians have a big red button hidden under their desks called ‘Analyse’ that will return the data they want in the format they want. Explain clearly the limits of what bioinformatics can achieve and how much work it actually is going to be for you to do what they’re asking. Let them know when their data is not suitable for rigorous analysis and why. Suggest solutions to these kinds of issues for their future work. This also maintains the position of you having intellectual input into the project, and the benefits that will bring.

Maintain communication

All too often interactions that start promisingly can dwindle to inactivity as both you and your biologists get sidetracked by other things. If you’re serious about the work you do, it is up to you to keep the interaction going. Personal organisation is the key here. Tracking your projects effectively, and knowing when to gently prod the biologists for the next round of interactions is all important. Don’t rely on them to come to you, go to them.

I don’t mind the BBC covering science but..

from this article:

this quote:

“The results showed the gene expression – the generation of functional proteins by a gene – declined with age”

A cursory glance at the paper (published in PLoS One and entitled “Resistance Exercise Reverses Aging in Human Skeletal Muscle”) shows that real time PCR was used to measure the gene expression of 500 or so marker genes for ageing.

Now, it’s called ‘gene expression’ because it measures the ‘expression level’ of genes. Genes are not proteins, and there is not always an obvious correlation between gene expression and protein production. I wouldn’t mind so much if they had not attempted to clarify this, rather than do so incorrectly.

I don’t know whether I’m abnormally fussy because most of my working life is spent staring at gene expression data, but I do wish sometimes that the science stories coming out of the BBC had actually passed by the eyes of a scientist before being published to catch some of the more obvious errors. Mind you I suspect the BBC’s commitment to science is always somewhat lacking .

Good news everyone!

Well good news for me.

I have in my hand a piece of paper that says “I am pleased to be able to offer you an open-ended contract with immediate effect”. This is undoubtedly “A good thing”™

And today, I have also become (thanks to the Arthritis Research Council) a PhD supervisor! This nearly happened a couple of years ago, but funding issues laid waste to those plans. This time the funding has arrived, due to an excellent proposal from my supervisee (Arthur), and I can’t wait to get into this role!

Is there no end to human laziness?

I think today I have finally experienced the zenith of consumer idleness.

As I was queueing at the cashiers in Tesco, happily perusing the contents of someone else’s shop (it’s pretty easy to feel virtuous about your shopping habits in Gateshead, as frequently you’re the only person who has bought an actual vegetable) I noticed a product I have never seen before, I don’t even know where you find it on the shelves in the supermarket.

It was a packet of “Tesco Healthy Living Baked Potatoes With Cheese”, packed in a plastic tray, in a plastic wrapper, with a plasticated cardboard cover.

Now, please, please, please can someone tell me why buying something like this, which (and I’ve checked Tesco online) is NOT microwaveable and must be cooked in the oven is preferable to buying a) a potato and b) some cheese and actually cutting the potato yourself and adding your own cheese?

Surely unpacking the potatoes from their plastic prison is practically as much work as slicing a single line through a soft baked potato and stuffing in a bit of cheese (one more slice of the knife)? Even at “2 for £3” i.e. 4 potatoes for £3, or 75p per potato it still must actually be cheaper to buy a) a potato and b) some cheese.

Given the packaging, the energy expended thereof, the inflated price and no doubt substandard quality of the cheese involved – at what point does an even vaguely rational human pick one of these up?

We’re all doomed.

Bank charges III

So a stunning silence ensued from NatWest after I politely approached them for repayment of their ill gotten gains. This was expected and merely requires that I escalate the game a little.

Today I stuck in the post (recorded delivery) my first ever threat of legal action against a company! It felt good, and indeed justified. Of course in 14 days or so I actually have to back this up by filing the court papers, but I’m convinced that it will be worth the effort to recoup my lost sterling. The “Letter Before Action” will drop on the doorstep of their Borehamwood Customer Services department and the clock will start ticking…

Oh the idleness of youth

A particularly stupid question from a budding bioinformatician on #bioinformatics on freenode

Names changed to protect the guilty.

[biofinformatics_student] i go to a school that has an account with most popular databases, but none seem to give access any earlier than 1992… sorry for such a general/stupid question but is there any way for me to find the text of a 1983 article?

[bioinformatics_student] the journal in this case is nature, and it’s not exactly cheap to pay the per-article fee

Clearly they don’t have libraries at Universities anymore…