Forum:F*up Night style events for Bioinformatics ? Comment if you're interested!

Question

[Ann]

Hi!

Have you ever made such an embarrassing bioinformatics mistake that you never dared to tell a soul about -- except maybe over drinks or other intoxicating substance?

Here are a few I've noticed over the years!

• You (or your sequencing facility) accidentally switched sample names (yikes - not my fault!)
• You forgot that a data file has a header row (oops, my bad)
• You forgot that R indexes start at 1 but python starts with 0 (oh no! sorry!)

Data analysis and data processing mistakes make you feel sick when they happen...but in the end, you're usually grateful you caught them in time to avoid even more embarrassment. But sometimes you don't!

Either way, after enough time has passed, it's fun to talk about and hear about big mistakes. The more we learn about the ways bioinformatics can go horribly, horribly wrong, the better we get at avoiding and recovering from mistakes.

Is anybody interested in attending (via zoom or in-person at, let's say, ISMB 2024 in Montréal) a f*up nights style event for bioinformatics?

If yes, and if you have some ideas for an event like this, please post in the comments!

Thanks a lot all!

• Ann Loraine

PS I'm a professor and I'm definitely not perfect.

[Pierre Lindenbaum]

11 years ago:

What Are The Most Common Stupid Mistakes In Bioinformatics?

[Istvan Albert]

Some things never change ... the most common mistake is still the one on top there

[Mensur Dlakic]

Pretty sure I have done it, but can't remember the details. You know that thing about blocking out traumatic events ...

[Madelaine Gogol]

That sounds fun, and I've got some of these that I have told many people about (to keep them from making the same mistakes as me). I am planning to go to ISMB and that sounds much better than zoom. : D

Answer 1

[Pierre Lindenbaum]

... reminds me https://twitter.com/yokofakun/status/1267877851795259397 ( my twitter account is now private)

Answer 2

[Dave Carlson]

I've definitely ruined some genome fasta files by running

grep ^> mygenome.fasta

instead of

grep "^>" mygenome.fasta

[Istvan Albert]

... ha, beat me to it, posted at the same time ... this error rules them all ..

here is a cool trick on how to count fasta headers in file only to first destroy your fasta file before it even executes

# Correct version
grep '>' mybigfastafile.fa | wc -l 

# How to destroy the file first
grep > mybigfastafile.fa | wc -l

and serves as an additional reason (among the many) of why one should always write command lines via the so-called useless cat streaming like so:

cat mybigfastafile.fa | grep ">" | wc -l

not so useless after all since it avoids a whole slew of errors

[Dave Carlson]

haha yes exactly! After the first or second ruined fasta, I started "catting" all fasta files and piping to other commands, just to reduce my stress levels.

Answer 3

[Mensur Dlakic]

I was trying to avoid re-living traumatic events, but here goes. This is an attempt to delete all the files with a certain extension but inadvertently adding a space after the asterisk:

rm * .csv

Not only will that delete all the files in the current directory, but it will further rub it in:

rm: cannot remove '.csv': No such file or directory

[Dave Carlson]

Ugh. I've definitely done this one before as well. And was none too happy about it.