Talk:In silico

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Derivation of in silico[edit]

As far as I can tell, silicate comes from the Latin silex meaning flint. It is fairly clear that flint is inferior to silicon as a description of a computer. Am adding something. Zargulon 18:21, 3 September 2005 (UTC)

No, sorry, that was wrong. Silicate also comes from silicium. Correcting ... Zargulon 18:43, 3 September 2005 (UTC)

History of in silico[edit]

I am Pedro Miramontes. I can prove that I used the expression "in silico" well before 1989. I did not include the references because they are in Spanish and they are not interesting for scientists outside Latin America or Spain. Nonetheless, my participation in NM in 1989 is recorded. Pedron 22:19, 28 March 2006 (UTC)

If there are (published) sources, please correct the article and cite those publications. Foreign-language publications are better than no sources, and I suspect that Spanish scientific journals are more authoritative than many English sources! Roberdin (talk) 15:41, 22 October 2011 (UTC)


I found this page (in silico) from a link on the mathematical biology page. I think that this article should be categorized in the mathematical biology section. —The preceding unsigned comment was added by InAJar (talkcontribs) 15:50, 28 February 2007 (UTC).

Image of DNA[edit]

The use of the DNA image in this article is pointless. I am removing it. Please, if someone puts it back give some rationale. Miguel Andrade 18:36, 7 March 2007 (UTC)

in italics[edit]

Why is in silico in italics?

  • Latin phrases in science are almost always put in italics. It's an incredibly common format, and you would be corrected if you fail to do so in a paper or thesis. —Preceding unsigned comment added by (talk) 20:03, 29 July 2009 (UTC)

Article Contents[edit]

Really? an article about "In silico" with literally no references to pure fundamental computational chemistry, molecular physics, materials science, etc? all of the links are related to biochemical and biological topics. there is no mention of the fact that this term is MOST applicable to more fundamental fields of the physical sciences. Seriously, you have to blindly biased to think that this term, or the conceptual idea behind it, has any specific relevance to biochemistry MORE than to physical chemistry and related topics. it is completely the other way. for ever research group doing protein folding experiments there are about 1,000 research groups model gas phase chemistry (applicable to everything from a mass spectrometer to upper atmosphere chemistry), solid state (think: semiconductor band gap modelling), and all other phases of matter under the influence of all manner of conditions. if anything, the biochemical applications are an extremely unique niche. Without question, computational biochemistry has had less success at accurate prediction or correlation with experiment out of any form of computational (non relativistic) physics. (talk) 04:56, 31 March 2011 (UTC)

The origin of the term is biological, it only arose as a contrast to in vitro and in vivo, and refers to the relatively novel application of computers to biological problems. It is a biological term, but it's possible it has been appropriated by other fields that already used computing, and I do think there could be a mention of other physical sciences to cover this possibility. (talk) 15:18, 2 May 2012 (UTC)

Is there any agreed on way to represent DNA using numbers (base 4 numbers would work well)

Say something like

0 = A is for adenine 1 = G is for guanine 2 = C is for cytosine 3 = T is for thymine

I am set back when wikipedia references a "digital" representation of a DNA sequence, I click on the link and instead of numbers I get an image file of letters.

([just a dumb question)] 03:27, 23 Aug 2011 (UTC) — Preceding unsigned comment added by Sunwukongmonkeygod (talkcontribs)

I agree that this article could use some balance. If this article is missing content that you are knowledgeable about, it is only because people such as you have not yet added your knowledge. Please feel free to improve this article. (talk) 11:40, 31 March 2011 (UTC)
The IUPAC standard for DNA and RNA nucelotides is letters, as shown here Nucleic_acid_notation. I am unaware of any numerical notation, certainly one is not widely used, and all online databases use letters. (talk) 15:18, 2 May 2012 (UTC)

Though there is no standard numeric representation of a DNA (or RNA) sequence the concept of using numbers for certain DNA calculations is pretty common. For example;

1. How many unique 6 nucleotide long DNA sequences are there?

We consider the 6 nucleotide sequence to be equivalent to a 6 digit base four number (ATGC). The largest possible value is 4^6=4096. But there are some tweaks to the basic calculation. If the DNA is double stranded then every time you count a sequence you automatically get the sequence's complement. So the number of unique 6-base pair sequence is closer to 2000. But, sometime the complement is the same as the original so you don't get the complement for "free". (For example, AAATTT has itself as the complement.) In single stranded DNA and RNA there are no tweaks and the number of sequences is just 4^n where n is the sequence length.

2. How long does a nucleotide sequence have to be for it to be expected to occur, on average, just once in the human genome?

Assume the human genome is 3x10^9 BP. We can start with 4^x=3x10^9 (and ignore the problem with complements) where x is the length needed to be expected to occur just once. Solving for x we get 15.74. So a reasonable estimate is, assuming DNA is a completely random sequence, that any 16 base sequence would be expected to occur about once in the human genome. (Real DNA sequences are not random.)

3. An average protein sequence is, say, 200 amino acids long. How many unique protein sequences are there of length 200? Can we estimate the total size of the sequence space that includes all possible permutations of length 200?

Here we are using protein sequences so the number base is 20 (vs. 4 for DNA). The number of unique sequences of length 200 is 20^200=1.6x10^260. That is not only a very large number but it is a mind-bogglingly large number. The number of atoms in the visible universe is only ~10^60, which is basically zero compared to 10^260. — Preceding unsigned comment added by 2001:18E8:2:28A5:BD7C:91AF:33DD:DA8D (talk) 17:49, 8 August 2012 (UTC)

The article states that in Greek, "silicon" would take the form "silico". This would be true if "silicon" were a Greek word. In fact, the name for the element Si in Greek is "Πυρίτιο" as can be verified by looking up the word in the Greek Wikipedia. Since the term "in silico" is not Greek, it doesn#t need to imitate Greek syntax. Nikos Bonanos

In silico vs. in silicium: better in silicio[edit]

I am referring to the sentence in the article which states that: " In silico was briefly challenged by in silicium which is correct Latin for "in silicon" ". To my knowledge, since modern Latin term for silicon is silicium (2nd declension neutral name in its nominative singular case), then the correct Latin expression for “in silicon” should not be in silicium (accusative case), but in silicio (ablative case).Ekisbares (talk) 16:30, 29 January 2018 (UTC)