Stejneger's Stonechat (1 Viewer)

[Richard Klim]

Stonechats: BBRC

BBRC, 24 May 2014: Stonechats – how BBRC are dealing with the new taxonomy.

Species Information & RIACT: Dippers to Buntings.

[Btw, for completeness/consistency, the RIACT account for Saxicola torquatus rubicola should be updated to reflect the BOURC split of European Stonechat S rubicola.]

 

[l_raty]

BBRC, 24 May 2014: Stonechats – how BBRC are dealing with the new taxonomy.

Genetic studies have shown that whilst variegatus is not robustly genetically distinguishable from maurus, the east Asian form stejnegeri (‘Stejneger’s Stonechat’) is clearly distinct from at least these two forms

Unfortunately, this is incorrect. Among the 8 birds that Zink et al. 2009 called variegatus, 2 (25%) were absolutely not genetically distinguishable from stejnegeri. Thus, on present knowledge, stejnegeri cannot be ID'd based on mtDNA alone. You need to exclude a Caucasian bird on morphological grounds.

(Particularly if this bird is one that reached W Europe at the same time as an obvious male Caucasian, I would say...)

 

[l_raty]

This bird was collected at Landsort, Södermanland, a few tens of km S of Stockholm, Sweden, on 2 Oct 2008.
The specimen (NRM-20086760) was barcoded in 2011 (SWEBI015-11 in BOLD).
It has stejnegeri-like mtDNA.

Is this a known record?

 

[Markus Lagerqvist]

This bird was collected at Landsort, Södermanland, a few tens of km S of Stockholm, Sweden, on 2 Oct 2008.
The specimen (NRM-20086760) was barcoded (SWEBI015-11 in BOLD).
It has stejnegeri-like mtDNA.

Is this a known record?

Very interesting! It's only published as Siberian Stonechat, not at race level (Stejneger's not split in Sweden). There are no accepted records of stejnegeri.

 

[l_raty]

Five Chinese barcodes have been added to the stejnegeri BIN in the BOLD database: 2 from Qinghai, 1 from Gansu, 2 from Hebei. (Click on the "PDF tree (All members and a member of the nearest BIN)" link on the BIN page for details; more than what is shown there remains hidden for now.) Unless these are all migrants from farther north, this suggests przewalskii is in.

(Pratincola maura var. Przewalskii Pleske, 1889: based on three birds from "Atschan Darja, mont. Russki, Turkest. or." (I'm not fully clear where this lies; East Turkestan = Xinjiang?) and four birds from Gansu.)

 

[Richard Klim]

Five Chinese barcodes have been added to the stejnegeri BIN in the BOLD database: 2 from Qinghai, 1 from Gansu, 2 from Hebei. (Click on the "PDF tree (All members and a member of the nearest BIN)" link on the BIN page for details; more than what is shown there remains hidden for now.) Unless these are all migrants from farther north, this suggests przewalskii is in.

So, the scope is potentially widening: Saxicola przewalskii Przevalski's/Eastern Stonechat (including ssp stejnegeri)? Or could it even prove to be S indicus (including sspp przewalskii and stejnegeri)...?

 

[Richard Klim]

Atschan-Darja

(Pratincola maura var. Przewalskii Pleske, 1889: based on three birds from "Atschan Darja, mont. Russki, Turkest. or." (I'm not fully clear where this lies; East Turkestan = Xinjiang?)...

From this text, Atschan seems to be in the vicinity of Qiemo (= Qargan/Cherchen/Tschertschen), Xinjiang (38°05'N 85°33'E). Presumably Atschan (At Shan?) is a peak in the Altun Shan, and Atschan-darja is a stream/river descending from it.

 

[l_raty]

From this text, Atschan seems to be in the vicinity of Qiemo (= Qargan/Cherchen/Tschertschen), Xinjiang (38°05'N 85°33'E). Presumably Atschan (At Shan?) is a peak in the Altun Shan, and Atschan-darja is a stream/river descending from it.

Thanks Richard :t:

So, the scope is potentially widening: Saxicola przewalskii Przevalski's/Eastern Stonechat (including ssp stejnegeri)? Or could it even prove to be S indicus (including sspp przewalskii and stejnegeri)...?

What is available from the range of indicus is:

- A single cytochrome b sequence from Illera et al. 2008 [pdf], indeed labelled indicus.
- Six brand new sequences (cytb, nd2, myo, odc, g3pdh, rag1--but mainly cytb and nd2 are interesting: there are not enough other available sequences of the nuclear genes for phylogeographic inference) obtained by Price et al. 2014 [pdf], from a bird labelled "Saxicola torquatus", from Kipsung in Nepal. (But note that this bird's ID has a question mark the the AMNH specimen database--choose "Ornithology", then enter "Saxicola" as genus and "torquatus" as species to find it--; it's just a tissue sample.)

Illera et al.'s indicus clustered with their S. leucurus, in a position basal to (torquatus + [rubicola + maurus]), which is compatible with the position of the stejnegeri clade. (As only nd2 and cox1 have been sequenced from birds known to be definitely part of this clade, being really sure is not possible.) Interestingly, though, Price et al. 2014 also produced sequences from a S. leucurus, including a cytochrome b that was highly similar to Illera et al.'s sequence of this taxon, and a nd2 that, when combined with other, ia. Russian sequences, appears sister to the stejnegeri clade. IOW, Illera et al.'s indicus was close to two congruent leucurus, one of which, based on another gene, was close to stejnegeri. Again, this is consistent with Illera et al.'s indicus being part of the stejnegeri group, thus, even if still not a direct proof.
(That being said... As you know, Illera et al. initially recovered sequences similar to their indicus sequence from two rubicola as well. I can only hope that the problem they had there, was due to contaminations. The other possible, and much more disturbing explanation would be numts. Should their original "indicus-like rubicola" sequences happen to be numts, their indicus sequence would certainly be wrong as well...)

And now, for the funny part:
The Price et al. 2014 cytochrome b and nd2 sequences are unlike anything that has been produced to date. They place this bird sister to the (rubicola + maurus) clade...

 

[Nutcracker]

The other possible, and much more disturbing explanation would be numts. Should their original "indicus-like rubicola" sequences happen to be numts, their indicus sequence would certainly be wrong as well...)

What are 'numts', please?

 

[l_raty]

What are 'numts', please?

Nuclear copies of mitochondrial genes.
http://en.wikipedia.org/wiki/Numt
If you include a numt in a phylogenetic analysis among genuine mtDNA sequences, the numt will show you a fake "divergence point" that in fact corresponds to the creation of the nuclear copy.

 

[Ben Wielstra]

Numts that have been evolving long enough to end up in a weird position would almost certainly show a messed-up reading frame, non-synonymous substitutions and stop codons, did you check for that?

 

[l_raty]

Numts that have been evolving long enough to end up in a weird position would almost certainly show a messed-up reading frame, non-synonymous substitutions and stop codons, did you check for that?

The reading frame is ok and there are no indels. Coding sequences with a messed-up reading frame, or stop codons, are now usually marked as such in GenBank (they end up labelled, eg., "cytochrome b-like", not "cytochrome b", even if originally submitted as such). If this had been the case, I'd have been much more affirmative. However, this does not appear always to be the case, and there are descriptions of numts that could be entirely translated.

What makes me feel somewhat uncomfortable, is that I see suggestions of intermediacy in some sequences from other studies, that as far as I can tell did not include any far-eastern birds.

Eg, FJ657482, a Saxicola rubicola from Tieleman et al. 2009 [full text]: this sequence is strongly divergent (4%) from other typical rubicola sequences; most differences match the indicus sequence; the similarity to indicus increases towards the end of the sequence. If you do a BLAST search restricted to the first half of the sequence (say, a query subrange of 1-600), the closest match is rubicola; if you restrict the search to the second half (say, 600-1143), the closest match is indicus. Yet the paper says: "We studied stonechats from three different populations belonging to three different subspecies, from Europe (Austria, Saxicola torquata rubicola), Africa (Kenya, Saxicola torquata axillaris) and Asia (Kazakhstan, Saxicola torquata maura), in addition to European–African hybrids and European–Asian hybrids." None of these populations/subspecies is supposed to include birds with basal (stejnegeri-type) haplotypes. Thus if this is a contamination, what is the source?

 

[Ben Wielstra]

Tieleman did sequence the Cyt-b gene in fragments and pasted those together to obtain the full sequence, so if they mixed up fragments (contamination or mislabelling) that could explain the different sigmal in different parts (although you would say the authors should have noticed mismatches when combining the fragments). But not why these fragments cluster in weird places though.

 

[Richard Klim]

Martin Garner, Birding Frontiers, 29 Nov 2014: Stejneger's Stonechat.

 

[Richard Klim]

Mortega et al

Mortega, Horsburgh, Illera & Dawson (in press). Characterization of microsatellite markers for Saxicola species. Conserv Genet Res. [abstract & preview] [pdf]

 

[Richard Klim]

Mortega et al 2014

Mortega, Flinks & Helm 2014. Behavioural response of a migratory songbird to geographic variation in song and morphology. Front Zool 11(85). [abstract] [pdf]

[via BF Newsroom.]

 

[Bartgeier]

Van Doren et al.

Van Doren, Campagna, Helm, Illera, Lovette & Liedvogel (2017). Correlated patterns of genetic diversity and differentiation across an avian family. Pre-Print, posted January 2, 2017

Includes a nuclear phylogeny of Saxicola rubicola rubicola, S. r. hibernans, S. torquatus axillaris, S. maurus maurus and S. dacotiae dacotiae. No samples of stejnegeri.

This seems to be another case where the phylogenies derived from nDNA and mtDNA are in conflict: according to this study maurus is sister to the remaining taxa and S. dacotiae is sister to S. rubicola.

 

[Nutcracker]

This seems to be another case where the phylogenies derived from nDNA and mtDNA are in conflict: according to this study maurus is sister to the remaining taxa and S. dacotiae is sister to S. rubicola.

Possibly not; if the S. maurus specimen is actually a misidentified S. stejnegeri. If that was the case, then if I'm recalling rightly, it would match the mtDNA tree more closely.

 

[njlarsen]

The methods state that the actual phylogenetic analysis was done on pools of 49-56 individuals per taxon, and describe the geographic home of the "maurus" individuals as Kazakhstan. They correlated the sequence results per pool with a refined sequence coming from a single Kazakstan bird.

Niels

 

[Nutcracker]

The methods state that the actual phylogenetic analysis was done on pools of 49-56 individuals per taxon, and describe the geographic home of the "maurus" individuals as Kazakhstan. They correlated the sequence results per pool with a refined sequence coming from a single Kazakstan bird.

Niels

Yep; but remember Zink et al. (2009; Mol. Phyl. Evol. 52: 769–773) had a Kazakhstan specimen embedded in stejnegeri - that's why I raised this possibility :t: