"Where can I get the source-code?"

I like to ask this delightful question : What are the units on the axes ?

Several replies so far and not one of "why did you write your program in XXX, when YYY would have been better" :->

I jest. I think the smartest questions are based on an actual desire for knowledge as opposed to ones that attempt to boost the ego of the questioner and/or attempt to demean the speaker. Perhaps another question to pose to biostar is "what is the best ego popping reply that you've heard from a speaker".

I digress, some generic questions I ask are whether the is code is free & open source, and if so, where I can get it. If I think the presenter is well versed enough, I'll ask how it compares to other programs in the field. If I want to help I'll ask the more awkward questions privately after the talk.

While I never heard it asked I like the question:

What you’re doing is rather desperate. Wouldn’t you be better off doing some experiments?.

My favourite was: "So just to be clear, all of these software packages are open-source and freely-available elsewhere?"

It was put to the CEO of a bioinformatics company which packaged up free software on a CD and tried to sell it. They even offered the NCBI handbook, "conveniently formatted as a single PDF" for $35.

Known exceptions to your hypothesis please..

Well, I often ask: "Does it work?" and by that I mean "Does it show/represent/display the real biology?"

And I swear the range of answers I've received are really funny. One time someone was showing us this very cool new splice variant display, based on some computational assessment. I said, "Ok. But does it work? Can you show me a sample that shows your display with some well-characterized splice variants from a known gene?"

The answer (I kid you not): "I don't know."

My favourite question when analysis results are presented with nary a word on the procedural intricacies is : Did "YOU" really do this? (Of course, I dont ask that in a seminar, seems preposterous!) Obviously the implication being I would like to meet the smart guy who did these analyses for you, drew your heat maps etc.. and you think R is a letter of the english alphabet.

My favorite question: So you studied Bioinformatics? Well, I have a problem. Do you know the Windows program XYZ? It doesn't work!

"Have you run your code to test for arsenic in the samples?"

What version of Excel did you use?

• What's about reproducibility of the results (when using random data)?
• Too low number of test cases.
• Your tests are biased.

did you test your model in real life ?

My favorite question is: which are your controls or models? This is because I am curious about how bioinformaticians validate their predictions.

For example, yesterday I attended a seminar from one of the Ensembl Compara developers, and I asked him: which genes do you use as controls when your update your pipeline? I wondered whether there are genes for which there is so much literature that we can be 100% sure of the orthologs in each of the species in Ensembl. I think that the choice of good controls for a pipeline is the moment where a bioinformatician is closer to the biology he/she is studying.

Well, my to question will be based on what kind of bioinformatics seminar it is- Generic or Specific. Though you haven't mentioned the topic of seminar then I will formulate questions totally based on my personal experience.

Specific (Next generation sequencing seminar)

Questions

• Enchancement and technological changes in third generation sequencing. What Equipments, Statistical and computational approaches, and pacakges associated with it.
• How analysis, test and hypothesis are formulated

Generic (Bioiformatics seminar)

Questions

• What are the most important tools that Bioinformatician must learn.
• Programming approaches and trends.
• How to keep up to date with the trends?
• References, Journals, blogs, citation and discussion groups

You have done with XXX software/programme, how reliable the software you have used? Have you compared your data with other software/programme?

How do you visualize the data after processing? Anything new here??