Question: Communicating With Non-Computational Biologists
gravatar for Damian Kao
7.7 years ago by
Damian Kao15k
Damian Kao15k wrote:

A significant part of our job is to be able to communicate what we do with non-computational biologists in an effective way without confusing them with technical jargon. However, sometimes you need to explain the technical process because it might pertains heavily to the biology. What tips or tricks do you guys have for communicating your results with a non-computational biologist whose extent of bioinformatics knowledge is knowing how to blast against NCBI?

ADD COMMENTlink modified 7.7 years ago by Martin A Hansen3.0k • written 7.7 years ago by Damian Kao15k

I don't think "traditional" is a meaningful term. Perhaps "non-computational" is better?

ADD REPLYlink written 7.7 years ago by Neilfws48k
gravatar for Stefano Berri
7.7 years ago by
Stefano Berri4.1k
Cambridge, UK
Stefano Berri4.1k wrote:

I am a biologist, but I gradually moved to the computational side about 6 years ago.

When I have to explain something involving math, statistics, computer science or something a bit technical I try to do the following:

  • Be graphic. Explain things as they where objects, with shapes and colors. Use a whiteboard. Biologist, and in particular molecular biologists, can't see anythink they work with (proteins, DNA, enzimes, transcription factors...) and are somehow used to build up things in their mind and make concepts happen there. If you explain microarrays, describe the spot as "a little lawn of single stranded oligos" and draw three of them on a board.

  • Use actual small numbers in your examples. To explain the Hypergeometric you can say something like: "If you have 4 red marbles and 2 black marbles, it is relatively easy that, drawing 3, you get all the black ones. However, if you had 40 red and 20 black, it is much harder to get all the black by drawing 30 marbles". Do not use formulas. Don't say "if you had M red marbles and N black marbles" You lose them straight away

ADD COMMENTlink modified 7.7 years ago • written 7.7 years ago by Stefano Berri4.1k
gravatar for Qdjm
7.7 years ago by
Qdjm1.9k wrote:

Great question! Stefano's advice is excellent, I have a few things to add.

  1. Be respectful. NCBs are not dumb, don't talk down to them. I hope no one would do this consciously, but it is also important to avoid doing it subconsciously. Perceived lack of respect is a barrier to effective communication. I find it helpful to keep in mind that biology is a very competitive field, so NCBs at the faculty, lab head, or even PDF level have to be outstanding researchers which means that, in addition to everything else, they are very intelligent -- just not as well-trained in math or programming.

  2. Know your audience. Not all NCBs are the same, some have more technical background than others. Try to take advantage of what they already know. All NCBs know what BLAST and a T-test are, but many also know the difference between local and global alignment and what linear regression is. Establish what they already know and relate what you are telling them back to those concepts whenever possible.

  3. Avoid unnecessary details. Why are trying to tell them about technical process? How does it relate to biology? Answering those two questions would be a great way to start your presentation. Then only explain in detail the part of the process that is relevant to the biology, making reference back to why you are doing this in the first place.

  4. Watch how other biologists communicate with each other. Biology is not just CBs and NCBs, it is also biochemists and non-biochemists, geneticists and non-geneticists, biophysicists and non-biophysicists, etc. Each of those sub-fields has to communicate technical details to non-specialists. Good biologists are experts at communication, so pay attention not only to what they are saying but how they are saying it.

A few things I learned watching NCBs communicate: avoid using a specialised term whenever an English word or explanation will do and communicate why you are doing something before you communicate how. Sometimes you don't even need to communicate how. Also, there are technical terms that NCBs understand for many of the concepts that you want to communicate, unfortunately they aren't always the same ones that computational scientists use.

ADD COMMENTlink written 7.7 years ago by Qdjm1.9k
gravatar for Zev.Kronenberg
7.7 years ago by
United States
Zev.Kronenberg11k wrote:

I figure if I can't plot it in a way that makes sense I need to rethink my test.

Here are my key tools for communication:

latex <- formats text, tables and figures beautifully ggplot2 <- exceptionally useful for basic plots circos <- great for displaying genomic positions.

A huge issue in communication for bioinformatics is getting the correct information from the biologists. I keep a lab notebook with notes from my meetings. Saved me more than I few times when people claim they give me different information.

ADD COMMENTlink written 7.7 years ago by Zev.Kronenberg11k

+1 for keeping a notebook. this is a great tool for preventing selective amnesia.

ADD REPLYlink written 7.7 years ago by Casey Bergman18k
gravatar for Bart Aelterman
7.7 years ago by
Bart Aelterman110 wrote:

Sure, that is a daunting task! Certainly when you have to explain your NCB boss why it is taking you so long to run that program (as if it were like pushing a button... right!) because you can't configure the third party software because you lack certain dependencies...

My advice: do get technical. However, explain it clearly. Everything. Tell them what dependencies are and what is different between those two platforms. Your colleagues are usually not dumb, which helps. And yes, that will result in long conversations, and sometimes difficult because of a lack of patience from both sides. To overcome that latter issue, try to be as enthusiastic as you can about what you're explaining, it will help them to stay patient. And if you have made yourself clear, they won't ask you again the following time.

I think it is far better to have long but clear conversations than a brief communication without the certainty that your message came across correctly.

ADD COMMENTlink written 7.7 years ago by Bart Aelterman110
gravatar for Larry_Parnell
7.7 years ago by
Boston, MA USA
Larry_Parnell16k wrote:

Communication - conveying an idea or process as well as listening and engaging in the dialog - comes easier for some and is a challenge for others. So, for one, I would identify who in the bioinformatics group is best at communicating with the non-computational biologists (NCBs) in this partnership. At the same time, the bioinformatics group may prefer to communicate with some NCBs over others because these are the good, patient listeners.

Two, I prefer to use examples or an analogy to get the bioinformatics concepts and processes across to my colleagues. Analogies require a certain amount of creativity. Examples require knowledge of the literature so that you can say, "that achievement published in Journal of X where they showed Y to be important in disease Z was made possible, in part, by the use of bioinformatics in the following ways..."

Three, just as the NCBs will learn (or be taught by you) about bioinformatics, you also need to learn all of the intricacies (or at least a few key ones) of the biology in which they have delved. This is all about finding common ground, which is the core of any successful partnership.

ADD COMMENTlink modified 7.7 years ago • written 7.7 years ago by Larry_Parnell16k
gravatar for Martin A Hansen
7.7 years ago by
Martin A Hansen3.0k
Martin A Hansen3.0k wrote:

In my experience the best thing is to sit down together for an extended period and work on a real-life problem. Biologists are not good at formulating questions that are suitable for a computer approach, and are often not aware of the opportunities - and they are too busy to read up. On the other hand, computer guys are not always good at understanding the biology aspects of the problem. Sitting together, both parties will learn - and hopefully the problem will be dealt with. And the process should be a bit easier for future problems.

ADD COMMENTlink modified 7.7 years ago • written 7.7 years ago by Martin A Hansen3.0k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1435 users visited in the last hour