How Genetic studies reveal new relationships, species

[jhyatt]

Thanks Adam.

"Smaller difference in COI". Noting the other factors- range, flight period, etc. how the heck then would one determine Sp vs SSP considering COI?

If we use 2% as a safe range for species (noting that many are now less than that), what might be the difference for a ssp? Must be less than 1/3 of 1%, considering glaucus and appalachiensis.

I expect the answer is "it depends" and "the other factors" but then still....we can argue morphology all day, argue about distant populations, etc...even combining all the factors, are there ANY guidelines for Sp vs SSP now?

To me, DNA data is just another character to use in making what is fundamentally a subjective decision. The DNA % difference is neither stronger nor weaker evidence of speciation than morphology and biology. Species are biological facts, but the rest of classification, both higher and lower levels, is a human construct. At least, that's the way I look at the situation....

jh

[Chuck]

(John) that makes me wonder what the pros have been doing for two centuries, arguing about whether something is a species or subspecies.

While CO1 gives some insight- and clears up some of these issues, I'd hope there was SOME standard being set...or at least a baseline guideline.

I see a lot of stuff in BOLD BINs that are all roughly equal off the same branch, and presumably then all manner of taxa(?) could be named as species.

[adamcotton]

COI does not really give a reliable indication of species status, only a suggestion of probability. On the other hand a combination of various mitochondrial genes and importantly also a number of nuclear genes gives a much better indication of specificity ... but even then results can be unreliable in some cases.

Adam.

PS. I agree with John's comment just above. DNA characters are just that ... another category of characters. They should be taken into consideration alongside all other 'traditional' characters.

[Chuck]

Interesting paper on genetically differing populations of African Ciclids (fish) cohabiting the same waters. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4700518/

Take-aways: (1) populations are genetically isolated despite being capable of interbreeding because of mating preferences; (2) the same situation is suggested for western US stick insects of the genus Timena, Heliconius, Drosophila, Anopheles, and a slew of non-insect animals.

In the case of the African Ciclids, the authors suggest that speciation may have started only 500 - 1000 years ago.

[Chuck]

I've been uploading to BOLD.

A collaborator (who's a name-brand professional) uses GenBank. He says "which BOLD mines now and then."

Um...should I be uploading data to BOLD or genbank? What's the difference?

The one difference I can find is that I can't figure out how to search for COI sequences on Genbank.

Appreciate any insight.

[adamcotton]

One advantage of BOLD is they require(?) a photo of the sequenced specimen and locality data to be uploaded as well as the sequence. That means it is possible to confirm the identity of the specimen that the sequence came from. GenBank was adapted from medical use to a biological one, and unless things have changed they only require country of origin (useless for confirmation of taxon identity) and the scientific publication reference, so it is/was impossible to confirm the sequence identity, especially given that taxonomic hypotheses change and also sometimes the original specimen was misidentified by the geneticist who sequenced it (I have found instances of this!).

Adam.

[Chuck]

Thanks Adam! Very useful. Being a ground pounder forced into the genetics stuff, I want an image and collecting data readily available. So I'll continue with BOLD.

[Chuck]

Question on COI tree; I understand the value (or lack thereof) in COI for establishing species status. So maybe I'm reading into this too much. Please clarify for me.



Group A, I know, is good old ordinary Papilio glaucus.

Group B is a sub-population that's regionally restricted and morphologically distinctive.

What the heck is C? As I read the tree, it implies that this specimen (at least, presuming it's part of a population) is an archaic offshoot that hasn't evolved in a million years.

[adamcotton]

What the heck is C? As I read the tree, it implies that this specimen (at least, presuming it's part of a population) is an archaic offshoot that hasn't evolved in a million years.

No, it implies that 'C' MAY have separated from the glaucus lineage before all the other subgroups and evolved separately (rather than not evolving at all).

The problem is that a tree based on COI alone is often unreliable (but not always), and a combination of a number of mitochondrial and nuclear genes is likely to produce a much more reliable tree, with reliability generally increasing with the number of genes included.

Adam.

[Chuck]

it implies that 'C' MAY have separated from the glaucus lineage before all the other subgroups

Thanks Adam.

Given that COI tracks some of the MtDNA mutations, is it safe to say that the one specimen (at least) of a long line of generations experienced some level of mutation that it has retained, and not bred back into the glaucus population?

Whether any given mutations are significant or not, I wouldn't know, but I'm not sure it's material.

BUT if 'C' MAY have separated? What other options are there?

[Jshuey]

It could also imply that you got a crappy COI read on that specimen. Check and see how many positions are in the sequence.

john

[Chuck]

Mitochondrial DNA (MtDNA) is a foundation of current genetic taxonomy. It is not comprehensive, it is not the latest and greatest.

MtDNA, beside being a partial snapshot, has one critical flaw- it is only passed down by the females. None of the male MtDNA is passed down.

So let's look at this.


This is a basic heredity tree, showing how each individuals is a product of one male (red) and one female (green) parent. Various types of DNA are passed from each of the parents. You are a sum of your ancestors- sort of.




However, MtDNA is only passed down from the female mother (green.) For the father (red), when the MtDNA from generation X-1 is passed to generation X, that MtDNA "disappears"- it does not get to the generation X individual. NOTE important- but other DNA does, otherwise you'd look like your mother.

Below, all the yellow MtDNA disappears when passed to the next generation because the males do not pass along MtDNA; the green is all that's passed down to the individual marked "Gen 7"




From a more practical perspective, what does this mean? Below, the Gen 1 female passes the MtDNA to the next generation, and IF that individual is a female she too will pass along this MtDNA.

In this example, if the Gen 1 female (and only the female) is Papilio glaucus, then that glaucus MtDNA will continue down the line of females only and the Gen 7 (whether male or female) will have the same MtDNA as the Gen 1 female.

COI is the analysis of some MtDNA, so the heredity of any individual, based on the COI Bar Coding, tracks only mutations in the MtDNA that came along a direct path on the maternal line.

Below, let's say that the green line of females are glaucus; however, the Gen 7 individual also has (obviously) a male father, which has ancestors, and in this case every single one, male and female is Papilio canadensis (yellow.) COI Bar Coding though would concretely put this Gen 7 individual in the Papilio glaucus clade.



Note too, for the Gen 7 individual, which I've selected to be male, his MtDNA will not be passed down, so all of the MtDNA shown in this tree is at a dead end, so far as his offspring are concerned; it's like they never existed.

So what use is MtDNA? First, it's a pretty good indicator of mutations. Second, it's inexpensive to extract and analyze.

Note too in the above drawings, it's unlikely that two parents produce only one viable offspring, so in any given population the MtDNA will be passed around in that population.

Hope this helps.

[adamcotton]

From a more practical perspective, what does this mean? Below, the Gen 1 female passes the MtDNA to the next generation, and IF that individual is a female she too will pass along this MtDNA.

In this example, if the Gen 1 female (and only the female) is Papilio glaucus, then that glaucus MtDNA will continue down the line of females only and the Gen 7 (whether male or female) will have the same MtDNA as the Gen 1 female.

I am a little confused by Gen 1 and Gen 7 in the figures, unless 'Gen 7' actually means the 7th generation ancestor. The oldest 'parent' in the lineage should be at the top of the pyramid with the offspring at the bottom, but perhaps I am misinterpreting your unusual pyramid.

Note too, for the Gen 7 individual, which I've selected to be male, his MtDNA will not be passed down, so all of the MtDNA shown in this tree is at a dead end, so far as his offspring are concerned; it's like they never existed.

So what use is MtDNA? First, it's a pretty good indicator of mutations. Second, it's inexpensive to extract and analyze.

Note too in the above drawings, it's unlikely that two parents produce only one viable offspring, so in any given population the MtDNA will be passed around in that population.

The last sentence answers the question ... the original male at the top of your pyramid almost certainly had many sisters who passed on the same mtDNA as the male to other members of the population. The main problem with mtDNA is when there is interspecific hybridisation involved. A first generation hybrid and all offspring down the female lineage of that hybridisation event will carry the mtDNA of the female parent species of the original hybrid. This can be problematic in the event that a single specimen is sequenced, IF that specimen was the result of hybridisation in the past. Whole genome analysis (expensive) solves this potential problem, and it may be the only solution for the glaucus-group, which has clearly undergone hybridisation events.

Adam.

[Chuck]

I am a little confused by Gen 1 and Gen 7 in the figures, unless 'Gen 7' actually means the 7th generation ancestor. The oldest 'parent' in the lineage should be at the top of the pyramid with the offspring at the bottom, but perhaps I am misinterpreting your unusual pyramid.

Gen 7 is a singular individual. It's parents are directly below it; and six more generations below (Gen 1) you see it's great-great-great-great-great-great grandparents.

This demonstrates the number of individuals that donate DNA to a single individual, exactly like all of my ancestors who donated their DNA (except in some cases MtDNA.)

[adamcotton]

Ah, I suspected that might be what you meant, rather than the top of the tree being ancestor of all those at the bottom, with only the green line having the same mtDNA. It's just unusual to see trees drawn that way round.

Adam.

[Chuck]

Subspecies generally have a smaller difference in COI between them than between species, but it is often best to treat subspecies as visibly distinguishable from the nominotypical population in the large majority of specimens.

With all the genetics work being done, I've seen a lot of ssp being elevated to full species status. And this brings an interesting question.

In binomial nomenclature the identity of an insect (and all else) is limited to two names: genus and species. However, ICZN recognizes subspecies with "A scientific name added as a trinomen on the end of a bionomen is taken to indicate a subspecies. However, it stops there."

So what of those various regional "forms"? Suddenly when a ssp is elevated to sp, NOW it too can have ssp, so some of those forms- which are not- can be taxonomically recognized. This is not new, though it seems genetic analysis has drastically increased this situation.

I wonder too- is the old taxonomic architecture dead? The genus-sp-ssp tiers are each equivilent, at the same level. But genetic trees show that the branches and nodes are not all at the same level. Some may dead-end at species; some appear to go beyond ssp...some of these "forms" are morphologically and genetically different than the sp/ ssp than the next branch over is to itself.

How will ICZN recognize these distinct below-the-subspecies populations? Will ICZN become archaic, and ignored, while the geneticists publish these great branching trees? It seems so in some cases; case in point the well known freshwater angelfish, Pterophyllum: there are to this day only three recognized species and zero subspecies. But the fish nuts have names for these unrecognized "populations". Why haven't the professionals bothered to describe new species and subspecies? Is it because, in their world, they all know what they're talking about, so don't need ICZN's taxonomic guidance? Is discovery moving too fast for science?

How do we reconcile ICZN's taxonomic rules with recent discoveries? Will there continue to be named populations that we circulate, like a secret, because the authority won't recognize it?

[Cabintom]

I think that if a subspecies is now recognized as a species, any "forms" that then merit being raised to subspecies level can be done so while adhering to the ICZN code. Depending on how they were described, some of those "forms" can have their status revised (pulled out of synonymy, stat. rev. or whatnot), others will have unavailable names and therefore would need to be described under a new name.

I expect Adam will have corrections, but I believe this is the essence of it.

[adamcotton]

Yes, Cabintom's reply is exactly a very simple explanation of the situation.

Names that were previously treated as synonyms of a subspecies and USED as form names in a polymorphic species may be used as subspecies in the case that they were originally described as a species or subspecies (this includes names described as a form or variety before 1961 as long as they were originally described in a geographical sense rather than as a sexual or individual variation) but if they were described as an aberration or the original description clearly indicates the name is infrasubspecific (such as a different coloured individual given a name) then the name cannot currently be adopted with its original author/year as an available name. Thus it is important to examine the original description of each name, and for a separate geographical population the oldest available name becomes the valid subspecies name IF subspecific or specific separation is warranted.

However,

So what of those various regional "forms"? Suddenly when a ssp is elevated to sp, NOW it too can have ssp, so some of those forms- which are not- can be taxonomically recognized. This is not new, though it seems genetic analysis has drastically increased this situation.

Forms of the same subspecies may appear individually SLIGHTLY different at the individual gene level, in the same way as people with blue or brown eyes would, but this genetic difference would be minimal, and is just individual variation in a single population. Particularly in polymorphic mimetic butterfly species local selection pressure on individual appearance can have significant effects. For example in the P. glaucus group the presence or absence of a dark female form probably depends more on the presence or absence of the model (Battus philenor) but may also depend on other unclear factors where the model does not occur but the dark form is still present. For instance perhaps the predators move around and learn to avoid the dark distasteful butterfly elsewhere and then subsequently avoid any they meet among the P. glaucus in a different place and/or time. It is worth noting that this can also be a temporal difference ... the model may have been flying weeks or even months before the mimic, but the predators in the same place associate the colour with a nasty event and remember it.

How will ICZN recognize these distinct below-the-subspecies populations? Will ICZN become archaic, and ignored, while the geneticists publish these great branching trees?

Not at all. Bear in mind that the ICZN Code governs NOMENCLATURE, not taxonomy. The Code tells you how to treat names proposed by taxonomists and which names can and cannot be used. It is up to the taxonomists to apply the 'rules' in order to choose the valid name for any taxon THAT NEEDS A NAME, excluding names that do not conform to the 'rules' of the Code. Bear in mind that each branch of a tree is a 'taxon' but they don't ALL need scientific names.

Any names originally proposed below subspecies level (= infrasubspecific) are unavailable under the Code and cannot be used as valid names (see above), but if a taxonomist decides that a POPULATION is worthy of subspecies status but does not have an available name that can be used as the valid subspecies name the taxonomist will describe a new subspecies, thus providing an available name to use.

How do we reconcile ICZN's taxonomic rules with recent discoveries? Will there continue to be named populations that we circulate, like a secret, because the authority won't recognize it?

I repeat, the ICZN does not set ANY taxonomic 'rules', only regulations governing nomenclature, as I stated above. This is a slightly subtle issue to grasp, but there is a distinct difference between nomenclature (how animals are named) and taxonomy (how they are related to each other). The Code does not set rules about how to distinguish a species or subspecies, or indeed a genus or a family, it just governs the names that taxonomists should use once they have decided how to classify the animals they study.

Note that in the case of angelfish the different coloured ones in captive breeding are probably like comparing different breeds of dogs. A Chihuahua is genetically not different to a Great Dane and does not deserve subspecies status, so the different coloured angelfish are not actually subspecies. It is possible that different rivers contain individual colours, but ichthyologists do not separate subspecies as much as lepidopterists for various reasons which they have opinions about, and as they all interbreed they are clearly the same species. Similarly the pigeons around here can be a number of different colours (white, grey, mixed etc) but they are all the same taxon and they all interbreed. Different branches of Zoology have their own way of classifying populations, based on criteria relevant to the animals they study. Some recognise subspecies, others do not, usually with reasons based on the individual biology of the type of animal.

Adam.

[Chuck]

repeat, the ICZN does not set ANY taxonomic 'rules', only regulations governing nomenclature, as I stated above

I knew you were going to say that. But the recognition of taxonomy is via nomenclature.


it does generate taxonomic rules by not allowing named sub sub species. These unique populations can’t be described as subsubspecies. Until a ssp is elevated to full species. THEN they can be described as a ssp. Until that point they have no nomenclature rules or protection of their informal identifier.

[adamcotton]

These unique populations can’t be described as subsubspecies. Until a ssp is elevated to full species.

Chuck,

The whole point here is:
a) that anything below the rank of subspecies is outside the ICZN Code, and as such any names proposed for such entities (individual forms, aberrations, sexual morphs etc) are not available names
and
b) these individuals are not recognised as comprising 'unique populations' but as variations within a population (= species or subspecies).

If subsequently a researcher finds that in fact certain individual variations found in the same place represent two different taxa THAT DO NOT INTERBREED then there is a process to decide the correct names to apply to these two:
1. Find out which of these two populations the name-bearing type belongs to
2. Name the other population according to the ICZN Code regulations as a separate NEW SPECIES (not subspecies - note that two subspecies of the same species do not normally occur in the same place, except for in relatively rare hybrid zones where the ranges of two subspecies overlap).

Adam.