[westerman]

@danwiththeplan: Your question doesn't make much sense since in this forum thread N50 is also known as L50. Since they are one and the same then there is no 'Whaaa?' involved -- they both represent a length in bases. Now if you have a paper you are reading that uses L50 in a difference sense then give us a reference to the paper so we can see what they mean. ....

Having had my attention brought back to this thread and having started using stats.sh from BBtools a lot more, I do note that stats.sh uses N/L50 to designate the Number of fragments at or above the Length50 cutoff.

ABySS uses the terms N50 and n:N50 for the same information. I suspect that most programs use 'N50' for the length information and something else for the number information. However Brian's nomenclature in BBtools makes good sense as well.

[Brian Bushnell]

I maintain that using "L" to denote a number and "N" to denote a length are just wrong, and I don't know how that got started. So I always define L50 as a length and N50 as a number. But Quast, for example, does use "L50" to indicate the number of contigs and "N50" to denote a length.

[kmcarr]

I maintain that using "L" to denote a number and "N" to denote a length are just wrong, and I don't know how that got started...

Really wish SEQanswers had a mechanism to "like" comments.

[westerman]

I maintain that using "L" to denote a number and "N" to denote a length are just wrong, and I don't know how that got started. So I always define L50 as a length and N50 as a number. But Quast, for example, does use "L50" to indicate the number of contigs and "N50" to denote a length.

You'll just love to know that the latest version of abyss-fac has ditched 'n:N50' (for the number) and 'N50' (for the length) in favor of 'L50' (for the number) and keeping 'N50' (length). Yes opposite of what you maintain and what makes logical sense.

[kbradnam]

I maintain that using "L" to denote a number and "N" to denote a length are just wrong, and I don't know how that got started.

I believe the precedent for this is in the 2001 publication of the human genome paper (in Nature): http://www.nature.com/nature/journal.../409860a0.html

"We used a statistic called the ‘N50 length’, defined as the largest length L such that 50% of all nucleotides are contained in contigs of size at least L."

Many others started using N50 after this publication, but I'm assuming that L50 for length came afterwards. So we now have a confusing situation, but unless someone can come up with a pre-2001 citation for use of L50, I suggest people stick with N50 for length and L50 for number of sequences (confusing though it is).

[kbradnam]

I've updated my blog post about this to include new evidence which also supports that N50 came first, and was first used in 2001.

[renekat]

Clarification is stats.sh BBTools

I maintain that using "L" to denote a number and "N" to denote a length are just wrong, and I don't know how that got started. So I always define L50 as a length and N50 as a number. But Quast, for example, does use "L50" to indicate the number of contigs and "N50" to denote a length.

Hello,
I think that makes sense and just want to verify that in the BBTools stats.sh and statswrapper.sh N50 is for counts and L50 is for length. Is that correct?
Thanks so much,
René