Talk:Gene

Suggesting 2015 GA Review
Transcluded from Talk:Gene/Review

Definition of "gene" (again)

 * '' Continued from talk:Gene

There are two different definitions of "gene" in the text and this needs to be fixed. We're talking about the molecular gene and the definition used by knowledgeable scientists is that a gene is a DNA sequence that's transcribed to produce a functional RNA. That RNA could be mRNA or any one a of a number of noncoding RNAs. That's the definition described in the first paragraph and it's supported by several appropriate references.

In the second-last paragraph we are introduced to the idea that "The concept of gene continues to be refined as new phenomena are discovered" and some of those "new phenomena" are supposed to be regulatory sequences and exons and introns. But regulatory sequences have been known for almost 60 years and they are not considered to be a part of the gene as defined in the first paragraph. Introns have always been considered part of the molecular definition of a gene ever since they were discovered about 50 years ago.

Another so-called "new phenomenon" is functional noncoding RNA but that's not new and it doesn't change the definition of gene that's used in the first paragraph. Knowledgeable scientists have known about noncoding genes since the mid-1960s. The fact that some genes are made of RNA deserves to be mentioned in the first paragraph so I've inserted two sentences.

The so-called "new" definition described in the last paragraph is " a broad, modern working definition of a gene is any discrete locus of heritable, genomic sequence which affect an organism's traits by being expressed as a functional product or by regulation of gene expression." I don't agree with this definition. It is supported by two references written by people who thought that the old definition referred only to protein-coding genes. (One them is Elizabeth Pennisi - a very unreliable source.) They were wrong and we don't need to quote people who had a misconception about the real historical definition of a gene.

I propose that we delete the second-last paragraph of the lead. Genome42 (talk) 22:14, 22 February 2023 (UTC)


 * I'd be fine with deleting the second-last paragraph of the current lead, especially since there seems to be a section dedicated to different definitions. It's probably still worth noting in the lede section that there are alternative definitions of a 'gene' other than the one in the very first sentence.
 * It'd probably also be worth organising that Definitions section a bit more, since it's currently a bit of a list of quotes (e.g. moving the Functional definitions subsection over from Structure and function). I think it's also better to focus on the fundamental aspects of each definition rather than who exactly coined it in most cases outside of the History section. T.Shafee(Evo &#38; Evo)talk 06:52, 27 February 2023 (UTC)
 * @Evolution and evolvability: I appreciate your effort to clarify the discussion over different definitions of "gene" but I really don't think your edits are helpful. You deleted a specific reference to Dawkins but I think that's vey important since "The Selfish Gene" is one of the most widely read books on the subject and it contributes significantly to confusion about the meaning of the word "gene," especially in the context of an article that's mainly about the molecular gene.


 * Breaking the section into subsections seems (IMHO) to make the discussion disjointed since two of the subsections ("Inheritance" and "Selection") both refer to the Mendelian gene and this article isn't about the Mendelian gene. That's covered under Genetics. In addition, your description of the Mendelian gene and its connection to selection is adequately covered under Gene-centered view of evolution and I think we should be linking to other articles when they cover a topic correctly.


 * Also, you added something about synthetic genes that isn't appropriate. Artificial DNA segments that some people refer to as genes are not relevant. The sentence on "de novo" genes is also more confusing than enlightening because in order to actually qualify as a "de novo" gene, the sequence has to meet the acceptable definition. The edit doesn't add anything to the discussion.


 * The problems are compounded by another discussion further down in the article under "Functional definitions." That discussion conflicts with the one we are editing and that's going to cause a problem later on. (Do we really need to waste time on rare overlapping genes when there's a very good article on the subject?)


 * Genome42 (talk) 18:16, 27 February 2023 (UTC)
 * I see what you're saying, though I think there are ways to note dawkins's influence on the popular understanding that flow a bit better. It may even work well to state the molecular definition first in that section (since it's the more common usage) then the second part can mention the continued contemporary use of a modern Mendelian definition in certain circumstances (e.g. forward genetics).
 * Wikipedia typically avoids phrasing around "This article focuses on..." and "More thorough discussions of this version of a gene can be found in...". It is probably better to state something more like "in a molecular biology context the definition most commonly used is XYZ. The reader can then see that the majority of the page is discussing molecular biology (except the mendelian inheritance section), then "in a genetics context (particularly forward genetics and gene-centred evolution), a mendelian definition is still sometimes used XYZ". That way a reader can see those contexts in the same way without the editorialised voice.
 * Are four examples of definitions needed as a list in the section? Perhaps it could work better to state the consensus definition before the minor variations that exist around it and to note what particular differences those examples exemplify.
 * I agree that the Functional definitions section needs to move up into this one and get integrated in. The whole Definitions section should probably end up 500-800 words. Overlapping genes probably deserves a sentence rather than a whole subsection. T.Shafee(Evo &#38; Evo)talk 03:43, 2 March 2023 (UTC)
 * There's a lot of misinformation on the web and one of our goals should be to counter that by posting reliable information on Wikipedia. But that's not sufficient because in order to counter misinformation you also have to debunk it.
 * In this case, the myths that need correcting are that up until the genomics era scientists thought that protein-coding genes were the only kind of gene and they thought that all noncoding DNA was junk. You and I know that's not true but statements to that effect are very common in the scientific literature. We need to spend a bit of effort showing that the real scientific definition of a gene hasn't changed substantially in 50 years in spite of what one might have heard or read.
 * Whenever you do that, it will sound like editorializing to all those people who are being asked to re-evaluate their preconceived notions. I realize that the Wikipedia culture is usually opposed to making strong statements about what's true and what's not but that's something that we need to change because it's getting in the way of critical thinking.
 * We have another problem. There are a ton of articles about molecular biology and they often cover the same topics and they often conflict. Can you guess how many times the structure of DNA is discussed? We need to clean up this mess by concentrating on a few high quality articles that can be linked to. This is one of those articles. We shouldn't be afraid of linking to other high quality articles for more information, especially if the topic is too complicated to summarize.
 * Along those lines, there are separate articles on Gene structure, Structural gene, Gene product, Gene family, Gene redundancy, Regulator gene, Pseudogene, Gene desert, Non-coding RNA, and Conserved non-coding sequence. Many of these articles cover the same topics and they often don't agree. Many of them discuss genes but they don't use the same definition we use here. This is a problem.
 * The term "overlapping genes" is a problem. In the case of well-studied prokaryotic examples what we're actually talking about is overlapping coding regions (not genes) and the overlap is usually only a few nucleotides. I don't think it deserves much coverage in this article; besides, there's already an entire article on Overlapping gene and another on Nested gene.


 * Genome42 (talk) 20:10, 2 March 2023 (UTC)
 * I agree that work should start from this page (assuming consensus is reached) and work outwards to harmonise. If we decide to include more than one example of each major class of definition, a simplified but updated version of this table or similar could help. T.Shafee(Evo &#38; Evo)talk 02:32, 9 March 2023 (UTC)
 * Your link brings up an issue that’s really important. The authors claim that genes are currently (2017) defined as DNA sequences that specify a protein then makes the further case that the current definition conflicts with the discovery of alternative splicing.
 * I think that’s incorrect and I can document my claim by quoting numerous textbook over the past 50 years that have defined a gene in a way that includes noncoding genes such as those for ribosomal RNA and tRNA.
 * How do we deal with conflicts like this? Do we have to give credence to every scientist who makes incorrect, misleading, or controversial statements because that’s what the Wikipedia culture demands or should we concentrate on giving the general public the best consensus view of knowledgeable scientists? Genome42 (talk) 18:45, 9 March 2023 (UTC)
 * @Genome42 - Sorry for the late reply on this. In the case of "One Gene -> One Protein", it should definitely be mentioned as a potential (common?) misconception or oversimplification and the reasons listed/explained. If it was fair simplification at one time then that should probably be mentioned (a bit like the Bohr atomic model), but I'm note sure this is really the same.
 * In general, if something is an uncommon misconception, then it can be easily omitted (or only briefly mentioned) to avoid WP:FALSEBALANCE. Similarly, WP:FRINGE positions can just be omitted. Genuinely common misconceptions (popular press, obsolete model, counterintuitive situation, oversimplification, misconception from another field etc), should generally be mentioned but immediately corrected (e.g. the misconception orthogenesis/progressionism in Evolution). A summary table would only be useful for when there are multiple reasonable alternative definitions that are commonly used by experts in relevant fields where we're at least alerting readers that alternative defs exist that come at an issue from different angles.
 * Also, since we now have a Definitions section, I've moved the Functional definitions subsection into it. Much of that subsection is now a bit redundant, so most can probably be omitted as the section as a whole is refined and condensed. T.Shafee(Evo &#38; Evo)talk 03:26, 20 March 2023 (UTC)
 * Which misconception are you referring to? Is it the misconception that all genes encode proteins or the misconception that protein-coding genes can only encode a single kind of functional polypeptide chain? Genome42 (talk) 12:19, 20 March 2023 (UTC)
 * In this case, both. T.Shafee(Evo &#38; Evo)talk 03:55, 21 March 2023 (UTC)
 * Which misconception are you referring to? Is it the misconception that all genes encode proteins or the misconception that protein-coding genes can only encode a single kind of functional polypeptide chain? Genome42 (talk) 12:19, 20 March 2023 (UTC)
 * In this case, both. T.Shafee(Evo &#38; Evo)talk 03:55, 21 March 2023 (UTC)


 * There don't seem to any objections to deleting the second-last paragraph of the lead so I have removed it because there is an extended discussion of gene definition elsewhere in the article. Genome42 (talk) 14:50, 7 March 2023 (UTC)
 * I saw Evo&Evo's post on the WP:MOLBIO talk page about the definitions section and I took a crack at rewriting it with a focus on brevity while trying to address some of the concerns discussed above. You can find it in my userspace here. This condenses the definitions section from ~1500 words to ~200 words, so a lot of neat details are gone, but some can likely be migrated to the History section or their relevant main article (if they are not already there).
 * I don't see why the definitions section should be very long at all if the goal is to provide extra context to what is meant be either the Mandelian or molecular gene. Extra nuance, such as the definition proposed by the linked 2017 article above, is probably too technical for such a general article. &#8213; Synpath 01:47, 12 March 2023 (UTC)
 * Thanks for taking an interest. Here's how I see it. We have several different audiences. The "general audience" probably doesn't care very much about the exact definition so the long version just looks like history to them.
 * The science crowd consists of readers who are interested in science and have probably taken an undergraduate course in biology. They have been bombarded with information about genes and how the old concepts are completely wrong and need to be drastically revised in the genomics era. It's likely they have heard some version of the story that old fuddy-duddy scientists (like me) thought that all genes encoded proteins and we couldn't adjust to the new ideas coming out of ENCODE and Evelyn Fox Keller. The long version is intended to correct that misconception.
 * Then there's the Wikipedians who are anxious to edit articles like this by inserting short references to statements "proving" that a new definition of gene is required because of alternative splicing and noncoding RNAs (and other things). It will be easy for them to do this with the short version but the longer version will (I hope) be more resistant to attacks from other editors.
 * It's a shame that we have to think about ways of protecting accurate science from well-intentioned, but uniformed, Wikipedia editors but it's a fact of life in 2023. Genome42 (talk) 18:23, 12 March 2023 (UTC)
 * I definitely don't know enough about Wikipedia to come down with a strong opinion on this. My intuition says an encyclopedia should prioritize the general audience, especially with a topic like this with such a broad cultural impression. That's why I opted for writing a shorter definitions section in hopes of increased accessibility. Maybe that's only most appropriate for the lede.
 * Also, I just noticed the hatnote pointing to the dab page doesn't use the molecular gene definition. I'll move the dab page definition over to the hatnote. &#8213; Synpath 18:49, 13 March 2023 (UTC)

Mutation rate
I don’t think we need to discuss mutation rate in this article since it is covered elsewhere. But if we include it, we should at least get it right.

The overall DNA error rate per replication is about 10^-10 - it includes the DNA replication error rate of 10^-8 and the fact that 99% of these errors are repaired. That gives 0.3 mutations per haploid genome per replication.

The mutation rate per generation in humans is not the same as the mutation rate per replication. The two papers that are referenced refer to the per generation mutation rate (10^-8). Thus every newborn baby has about 60 (2 X 30) new mutations according to this mutation rate - the latest data is closer to 100 mutations per human generation. Genome42 (talk) 02:07, 5 December 2023 (UTC)


 * I strongly recommend that we delete the section on "Molecular evolution" since it doesn't belong here and the material is covered elsewhere. The material in the subsection on "Mutation," for example, is covered in Mutation rates where the mutation rate in humans is said to be 50-90 mutations per generation. This conflicts with the value of 30 that was just added to this article. (The Mutation article is closer to being correct and the value stated here is wrong. The actual value is probably closer to 100 but we'll deal with that in the Mutation article.)
 * This example illustrates the problem with redundancy. When the main article is updated and corrected, the other entries become wrong and this helps spread misinformation and confusion. Genome42 (talk) 21:14, 10 December 2023 (UTC)

Book "How Life Works" (2023) worth considering?
A review by scientist Denis Noble of a new book entitled "How Life Works: A User’s Guide to the New Biology" (2023) by Philip Ball (editor of the journal Nature) may be worth considering? - iac - Stay Safe and Healthy !! - Drbogdan (talk) 04:51, 6 February 2024 (UTC)


 * Nonsense. Denis Noble is not a credible authority on genes and neither is Philip Ball. I haven't got my copy of Phil's book yet but I'm familiar with his earlier writings. This is very controversial and bound to get us into bitter edit wars. Genome42 (talk) 18:25, 7 February 2024 (UTC)

Drbogdan (talk) 04:51, 6 February 2024 (UTC)