Freethought & Rationalism Archive

DNAunion · 03-30-2003, 01:28 PM

Quote:

Principia: No, DNAunion. I don't understand why you keep talking about heterodimers when I am specifically talking about homodipeptides.

DNAunion: Uhm, a great deal of my post dealt with Ala-Ala.

More importantly, the underlying logic I am explaining is the same whether we are talking about Ala-Ala or Ala-Asp or ....

In long chains of amino acids (not dipeptides) the amino acids linked to both sides of the two of interest can influence how readily those two amino acids of interest bond together.

For example, (L1) Leu-Ala and (R1) Ala-Asp may have a different "preference for bonding together" than (L2) Glu-Ala and (R2) Ala-Gly.

Why? (L1) has a fairly bulky, neutral amino acid (Leu) attached to Ala, whereas (L2) has a fairly bulky, negatively charged amino acid (Glu) attached to Ala. So the two "left sides" may interact differently with a "right side".

And, (R1) Ala-Asp has a moderate sized, negatively charged amino acid (Asp) attached to Ala, whereas (R2) Ala-Gly has a small, neutral amino acid (Gly). So the two "right sides" may interact differently with a "left side"

Combining these two possibilities suggests that Leu-Ala-Ala-Asp will occur more or less often than does Glu-Ala-Ala-Gly, even though both of them have Ala-Ala.

Principia · 03-30-2003, 01:34 PM

Alright, maybe one last comment:

Quote:

Combining these two possiblities suggests that Leu-Ala-Ala-Asp will occur more or less often than does Glu-Ala-Ala-Gly, even though both of them have Ala-Ala.

Quite clearly you have not read Rode's model of the mechanism behind SIPF. Nor did you understand his experimental protocol. How are we talking about Leu-Ala-Ala-Asp when neither Leu nor Asp were in the reaction? Rode is quite explicit in that he was dealing with a binary system at most, not quarternary, or ternary. In other words, either two different amino acids were present in a reaction or just one. Furthermore, the Ala-Ala values were reported for one reaction, in which there was only Ala present. The yields were not measured across all experiments. Read the paper

Once again, let's remind ourselves that my point here is in regards to your double counting of coincidences -- yet another flaw in addition to your biased uniform distribution.

Alright, now I know that I have to quit, when after 4 attempts by DNAunion, I am still repeating the same point. I am sure there will be a 5th, then a 6th, etc. ad nauseam.

DNAunion · 03-30-2003, 01:42 PM

Quote:

Principia: 7) So what this whole exercise amounts to is mere nitpicking.

DNAunion: You call Rode's being off by at least 11 orders of magntiude being nitpicky?

And yet you thump your chest in victory because the code I used to expose his flaw has a one-tenth of one percent "error bar"?

DNAunion · 03-30-2003, 01:53 PM

Quote:

Tom Ames: Um, a minor quibble: using <iostreams> technically makes your program C++. But other than replacing the "printf" statements with "cout", you've really written nothing that couldn't have been done in C.

DNAunion: Which is completely irrelevant. Principia showed his ignorance of progamming when he incorrectly refered to the language as C instead of C++.

Let�s see. What clues were there that the program was written in C++ and not C.

1) I explicitly told Principia it was C++. That alone should be enough!

Quote:

DNAunion: Perhaps it's just that you can't "speak" VFP: maybe you can only decipher a "real" programming language like C++. Well here, I took the time to recode it in that language.

2) The program extension is CPP � a clear indicator that it is a C++ program

Quote:

DNAunion:
// calcprob.cpp
// This program models choosing lettered tiles from an urn in
// order to calculate an empirical probability

3) I included IOSTREAM instead of STDIO.H

Quote:

DNAunion:
#include <iostream>

4) I am using .NET, as indicated by the use of a namespace: and Microsoft has a C++ .NET but not a C .NET

Quote:

DNAunion:
using namespace std;

5) I used multiple COUTs instead of PRINTFs

Quote:

DNAunion:
cout << "Tiles discarded after being chosen? ";
cout << (nDiscardTilesOnceChosen == 1?"Yes":"No") << endl;
cout << "Number of lettered tiles in Urn: ";
cout << nLetteredTiles << endl;
cout << "Number of target tiles: ";
cout << nTargetTiles << endl;
cout << "Number of matches needed: ";
cout << nMatchesNeededForSuccess << endl;
cout << "Trials per iteration: ";
cout << nTrialsPerIteration << endl;
cout << "Total iterations: ";
cout << lIterations << endl;
cout << "Successful iterations: ";
cout << lSuccessfulIterations << endl;
cout << "Empirical probability: ";
cout << ((float)lSuccessfulIterations / lIterations) * 100 << "%" << endl;

DNAunion: That�s plenty of clues for anyone familiar with the programming languages to figure out that it was C++ and not C. In fact, even people not familiar with programming could tell it was C++ BECAUSE I SAID IT WAS.

DNAunion · 03-30-2003, 02:00 PM

Quote:

DNAunion: I didn't see any GOTOs or other indicators of spaghetti code in my programs. Can you point them out for us?

Quote:

Principia: LOL. Every if statement you have is a branching code that does 2 things in the compiled code:

1) it does a test

2) it jumps to the right code (and provides a return address for when this code finishes)

DNAunion: Uhm, conditional branching using IF constructs is found in all non-trivial programs. "IF" is a standard construct used by all programmers, across the world, regardless of what language they are coding in.

Your ignorance of computer programming is showing every time you open your mouth! Please, keep talking!! You're making my job so easy! LOL!!!

DNAunion · 03-30-2003, 02:08 PM

Quote:

Principia: So three if-else if blocks, three implicit branch commands.

DNAunion: So? If a computer program never branches conditionally, it can�t �make choices�. That is, it can�t do anything �meaningful�.

Principia, show us how YOU would determine IF a tile had already been chosen WITHOUT using any conditional branching of any kind.

Come on, let's see some of your magic coding abilities! LOL!

pz · 03-30-2003, 02:13 PM

OK, guys, I see spelling flames and a lot of similar sniping over similarly inconsequential trivia. If you can't get over the need to score points on each other, I'm going to close the thread.

Principia · 03-30-2003, 02:20 PM

Quote:

Originally posted by pz
OK, guys, I see spelling flames and a lot of similar sniping over similarly inconsequential trivia. If you can't get over the need to score points on each other, I'm going to close the thread.

pz, I would rather you not, because the ensuing posts by DNAunion show clearly how he is hyping up irrelevant issues to dodge the main points of this thread, which as I have stated repeatedly are:
1) DNAunion has no grasp of Rode's experiment. His latest X-A-A-Y explanation clearly demonstrates that;
2) DNAunion has no grasp of Monte Carlo simulations, which he relied on to "debunk" Rode's work;
3) The difference between 1e-7 and 1e-18 is one of model. Technically speaking, Rode is absolutely right in his implementation (as exemplified in the OP thread). What he got wrong, and which DNAunion touched up, was the model -- that is to say the one he used was overly restrictive. Rode got the wrong model but the right calculation; DNAunion had the wrong model and the wrong implementation. Either way, they're both wrong, and imo, their numbers are just as irrelevant to the conclusions.

I'll happily grant what peripheral issues DNAunion is going to use to snipe at me for the remainder of this thread, since they don't even address the Rode paper.

Principia · 03-30-2003, 02:24 PM

Just can't seem to get away from this thread:

Quote:

Principia, show us how YOU would determine IF a tile had already been chosen WITHOUT using any conditional branching of any kind.

Come on, let's see some of your magic coding abilities! LOL!

Nah, I don't need no magic when I've got a pencil and paper. Plus, suppose I do fulfill another one of your requests, what will it show? Just look at the 5/14 calculation you asked me for. Dropped. Just look at the critique of your code you asked for. Ignored (or more accurately, it was turned into more taunting requests like the one above). Sorry, as I said before, I don't jump hoops for you.

Principia · 03-30-2003, 02:30 PM

Quote:

Principia, show us how YOU would determine IF a tile had already been chosen WITHOUT using any conditional branching of any kind.

Easy. I wouldn't do a Monte Carlo analysis, since I already have the exact formula.

03-30-2003, 02:13 PM	#67
pz Veteran Member Join Date: Aug 2001 Location: Morris, MN Posts: 3,341	One bucket of cold water, right here. OK, guys, I see spelling flames and a lot of similar sniping over similarly inconsequential trivia. If you can't get over the need to score points on each other, I'm going to close the thread.

Thread Tools	Search this Thread
Show Printable Version	Search this Thread: Advanced Search

Freethought & Rationalism Archive

The archives are read only.