Question

base of B-value log in Limma

0

Entering edit mode

Richard Friedman ★ 2.0k

@richard-friedman-513

Last seen 9.6 years ago

Dear Bioconductor users, I have been assuming that the base of the B-value log in Limma = 2, but I recently reread the paper and didn;t see a base. Is the base 2, and if not what is it/ Thanks and best wishes, Rich ------------------------------------------------------------ Richard A. Friedman, PhD Associate Research Scientist Herbert Irving Comprehensive Cancer Center Oncoinformatics Core Lecturer Department of Biomedical Informatics Box 95, Room 130BB or P&S 1-420C Columbia University Medical Center 630 W. 168th St. New York, NY 10032 (212)305-6901 (5-6901) (voice) friedman at cancercenter.columbia.edu http://cancercenter.columbia.edu/~friedman/ "Cartesian duelism is when somebody told Decartes that he didn't think therefore he was, and Descartes challenged him to a duel". -Isaac Friedman, age 16

Cancer limma Cancer limma • 1.4k views

ADD COMMENT • link updated 6.8 years ago by Gordon Smyth 50k • written 18.0 years ago by Richard Friedman ★ 2.0k

Gordon Smyth · Answer 1 · 2006-05-04

0

Entering edit mode

Morten ▴ 300

@morten-929

Last seen 9.6 years ago

Hi
From Replicated Microarray Data, Lonnsted & Speed it seems to be the "natural" log with base 10. (Equation 3 in the paper.) I presume this is implemented in Limma.
There is no reason to belive the base is 2.
HTH
morten

ADD COMMENT • link updated 6.8 years ago by Gordon Smyth 50k • written 18.0 years ago by Morten ▴ 300

0

Entering edit mode

Dear Morten,

Thank you for your answer, but I am still not sure based upon it. In Limma log2 is used for absolute intensities, and for M. M=log2(Intenisity2)-log2(Intensity1). In limma, B and M appear in the same table and graphs. This is unusual if B and M have different bases.

Also, at least in the usage with which I am familiar, "natural: logs are base e not base 10" and I have seen "log" without a subscript used for both base 10 and base e logs. So even if B is not base 2 I am not usr eif it is base 10 or base e.

Would Gordon care to comment?

Thanks and best wishes,
Rich

ADD REPLY • link updated 6.8 years ago by Gordon Smyth 50k • written 18.0 years ago by Richard Friedman ★ 2.0k

0

Entering edit mode

Quoting Richard Friedman <friedman at="" cancercenter.columbia.edu="">: > Dear Morten, > > Thank you for your answer, but I am still not sure based upon it. > In Limma log2 is used for absolute intensities, and for M. > M=log2(Intenisity2)-log2(Intensity1). In limma, B and M appear in > the same table and graphs. This is unusual if B and M have different > bases. Why is it unusual? we use log2 for the ratios simply so that for each "unit" of M we double the ratio, as that seemed a reasonable scale. You could use log5 if the scale of change justified it, for your purposes. A while ago I was curious about the B value calculation and how the probabilities were translated into B, and I noticed that the log was definitely not in base 2. But I can't remember if it was base 10 or e... If noone gives an answer I'll fish out my notes... Jose -- Dr. Jose I. de las Heras Email: J.delasHeras at ed.ac.uk The Wellcome Trust Centre for Cell Biology Phone: +44 (0)131 6513374 Institute for Cell & Molecular Biology Fax: +44 (0)131 6507360 Swann Building, Mayfield Road University of Edinburgh Edinburgh EH9 3JR UK

ADD REPLY • link 18.0 years ago J.delasHeras@ed.ac.uk ★ 1.9k

score 0 · Answer 2 · 2017-07-13

The B-statistic is on the natural log scale (which means base e). In documenting, I followed the usual R practice that logs are always natural unless otherwise stated.

Lonnstedt and Speed (2002) used log10, but natural log is more usual for this sort of statistic in other fields of statistics.

There is no reason why logFC and B should use the same log base as they are quite different sort of quantities. The concept of doubling a log-odds quantity has no particular intuitive meaning whereas it does for a log-fold-change.