my home is in empty spaces.

Suppose that 20 units of the fluorescence observed is due to background fluorescence, 32 units is due to background fluorescence + water, and 68 units is due to background fluorescence + water + drug of interest.

20 background fluorescence
32 background fluorescence + water
68 background fluorescence + water + drug of interest

And we want the fold change in fluorescence due to the drug. Would it be (68/32)= 2.125-fold, (68-32)/(32)= 1.125-fold, or (68-20)/(32-20)= 4-fold?


Edit 1: No, I am not working with microarrays. This is not a microarray question. Any insights from microarray people would be appreciated though.

Edit 2: I like this answer here, supporting the 2.125-fold calculation. Whether it's correct is another matter.

"Associative arrays are usually used when lookup is the most frequent operation. For this reason, implementations are usually designed to allow speedy lookup, at the expense of slower insertion and a larger storage footprint than other data structures (such as association lists)." (wikipedia)

Having a very small child to chatter at, wave brightly-coloured things at, and bring sightseeing around his present domicile, makes me wonder how to distill the meaning of life into choosing what to chatter to him about, wave in his line of vision, and et cetera. Most of these heavy decisions will be left to his parents =) . Perhaps for all of his guiding period I will continue as I have begun: work with my family and his parents to assist him in becoming the best person he wants to be. Necessarily this most involves listening to his views on what goals he wants to have, regardless of how apparently silly or inappropriate they are. Already before he was born, I was fully committed to the idea that he might be a graphic designer or something equally out of the realm of all our combined experience. It would be an interesting challenge. However at present he merely wants to turn over. =)

Friend: i think it's easier for memories to retain the clarity of the initial experience if the frames of reference of the initial experience have not shifted around them
Me: ah.
Me: a good way of saying it
Friend: if you or your life changes, memories formed pre-change start to lose definition
Me: indeed.
Me: and i recover by changing quickly, to admit the possibility that yes "it" not only exists but has happened
Me: and has (had........) in fact happened to me .......
Me: a very fast change process, i can do.

Six months, nearly.

The faintest twitch of memory leads me to check back my records. Not six months, as it turns out, but 1 1/2 years.

One-and-a-half years.

A wound, bone-white flesh with faintest threads of pink, little blood welling out, neither hot nor cool to the touch. No pain, not infected. Innocuous. Perhaps, healing. But not healed, as a recent event has shown.

You would not know the blood that gushed from it upon your thoughtless words. It is your purest luck that you do not know. No, I have no intention of ever understanding myself fully. There are things that man is blessed with, one of them is the ability to forget. But I had by then forgotten how to staunch this wound when you opened it.

In this respect I have forgotten more than you have ever known.

I intend not to speak with you again.

Cryptic note:

Cluster concept, cluster concept *mutters*. Singly not necessary and jointly oversufficient. What is a lemon? What is a game?

God's Debris: a thought experiment, by Scott Adams (creator of Dilbert). I'll find the time someday to read it.

Link via thegreatsze.

Ronald N. Giere, in Taking the naturalistic turn: how real philosophy of science is done, UChicago Press 1993, Werner Callebaut ed., p. 317:

"Science is a complex phenomenon. It has all kinds of aspects that one could not expect to capture in a single family of models - even a fairly extensive family. There are cognitive aspects involving the creation of models and judging how well a model fits the world. There are social aspects involving communication and other forms of interaction among scientists. And there are aspects of change, for which an evolutionary model may be quite appropriate. I think an adequate theory of science will require at least these three sorts of models. The problem will be to get them to mesh comfortably together."

The following are from a textbook by Arthur Lesk, Introduction to Protein Architecture (OUP 2001). Page numbers are indicated.

Unnumbered table. Structural parameters for protein secondary structures. (pg 67)

Structure                      phi    psi       n     d    p 
alpha-helix                   -57    -47       3.6   1.5   5.5
3(10) helix                   -49    -26       3.0   2.0   6.0
pi-helix                      -57    -70       4.4   1.1   5.0
Polyproline II helix          -79    +149      3.0   3.1   9.4
// beta strand                -119   +113      2.0   3.2   6.4
anti-// beta strand           -139   +135      2.0   3.4   6.8

n = number of residues per turn
d = displacement between successive residues
p = the pitch of the helix, the distance along the helix axis of a complete turn.

p = n times d. (The equation is exact; the values of p, n, and d in the table have been rounded to 2 sig. fig.)


Unnumbered table. Sequence-structure relationships in two-residue beta-hairpins. (pg 97)

Sequence    Conformation                  Type
XGXX        beta-epsilon-gamma-beta       II'
XXGX        beta-alpha(L)-alpha(L)-beta   I'
XXXG        beta-alpha-alpha-epsilon      I
XXXX        beta-alpha(L)-gamma-beta      III'
XGGX        beta-alpha(L)-alpha(L)-beta   I'
other       beta-alpha(L)-gamma-beta      III'

Useful for predictions of short beta-hairpins up to 6 residues in length. The conformation notation here is from Efimov AV, Mol Biol (Mosk). 1986 Jan-Feb;20(1):250-60. [Article in Russian. Image of alpha, alpha(L), beta, delta, gamma, and epsilon regions is reproduced as Fig. 3.17 in Lesk, somewhat resembling a Ramachandran plot.]


Text. The structure of helix-helix packings. (pg 141-143)

The relative geometry of the helices can be described by the distance of closest approach between their axes, and the interaxial angle. Typically, interaxial distances are 6-10 A, with ~ 1 A interpenetration of the sidechains. In order to achieve good packing densities, the two interfaces have complementary surfaces, like the occluding surfaces of teeth.

...

[Since the alpha-helix is 3.6 residues per turn], residues separated by four in the sequence are close together on the helix surface. Sidechains at these positions are poised naturally to create ridges, [...] in an ideal alpha-helix, the line joining the Cbeta of residue i to the Cbeta of residues i+4 and i-4 will make an angle of 26 deg with the helix axis. It follows that forming an interface between these ridges on the two helices will fix the interaxial angle at a value around -52 deg. (The negative sign follows the convention that a clockwise rotation is negative.) This is indeed near the average for a commonly-observed class of interaxial angles.

Interacting ridges i+-4/i+-4 i+-3/i+-4 i+-1/i+-4
Interaxial angle    -52 deg   +23 deg  -105 deg

Although most helix interfaces fit the 'ridges-into-grooves' model, there are exceptions. For example, in the B-E helix packing in globins, ridges from either helix cross each other, at a notch formed at a pair of glycine residues.


Unnumbered table. Geometric parameters of beta-barrels. (pg 154)

n = number of strands
S = shear number, the difference along the sequence of the residues forced to correspond when the barrel is closed by superposing the two copies of the first strand
alpha = angle of tilt between strands and barrel axis
?? = twist of the strands, the average angle between adjacent strands
R = radius of barrel

a = Calpha-Calpha distance along strand = 3.3 A
b = perpendicular distance between strands = 4.4 A

R = [(Sa)^2 + (nb)^2]^1/2 / [2n sin (pi/n)]
tan alpha = Sa / nb