Page 12 of 23 FirstFirst ... 2101112131422 ... LastLast
Results 111 to 120 of 222

Thread: Ancient I-M253 samples list

  1. #111
    Registered Users
    Posts
    379
    Sex
    Location
    United Kingdom
    Ethnicity
    European
    Nationality
    British
    Y-DNA
    I-L338
    mtDNA
    J1c1

    United Kingdom England England North of England Norfolk Scotland Ireland
    Happy New Year fellow ancient I-M253 enthusiasts.

    It's been a while since I posted anything on this thread. We've discussed earlier on this thread the two I1 samples CL63 and SZ45 from the "Understanding 6th-century barbarian social organization and migration through paleogenomics" paper that was published in Nature Communications in 2018 by Amorim et al. Link to the paper and supporting information since it's been a while (although I'm sure most people reading are familiar): https://www.nature.com/articles/s414...-06024-4#Sec23

    Specifically regarding SZ45, which was one of the samples that had a WGS and was listed as I1a1b1 (I-L22) in Supplementary Table 2. Looking at Supplementary Data 3, there's an excel spreadsheet summarises the Y-SNP analysis. While they list a few of the I1 SNPs, and record a positive result for L22 (which agrees with I1a1b1), they also list a positive result for Z63. We've mentioned this before on this thread, and we know that one of them has to be wrong. They also list P109 in among the I2 SNPs, so I think we can agree that the raw data is worth looking into further to clear up some of these discrepancies and get to the bottom of this.

    The BAM file for SZ45 is available from NCBI, so I downloaded the file and took a look using the Broad Institute's IGV. SZ45 is clearly negative for Z63 (11C reads in the BAM), and L22 has a single positive read. Single positives can be false positives, but we also have good multi-read results for Z2337+ (6A) and Y3549+ (5T) upstream and downstream which adds weight that L22+ is genuine. Working further down the tree and eliminating branches based on negative results and confirming based on positive, I'm able to find P109+ (6T) and S7660+ (5A).

    I'm pretty confident that SZ45 would be at least I-Y14999 on the YFull tree https://www.yfull.com/tree/I-Y14999/ and I-Y15520 on the FTDNA public haplotree. There are a couple of SNPs that are ambiguous reads, namely FGC21822 (4C, 1T) and BY35337 (10G, 3T) and another one has only one positive read (FGC21805). However, nine phyloequivalent SNPs on that branch are all multi-read positive - these are Y15520+, Y14999+, FGC21824+, FGC21823+, FGC21813+, FGC21814+, FGC21807+, FGC21811+, FGC21802+. This all adds up pretty well for this branch.

    I can rule out a couple of the known descendant subclades that are on the YFull tree and the FTDNA public haplotree. Well characterized calls for BY35334-, BY42214- and BY78500- on one branch; and for the other branch Y37415-, BY35335- (DYZ19 region), Y37108-, Y37417- and Y37416- on the other branch.

    SZ45 may be part of the I-FGC21819 branch, and I think he probably is FGC21819+. However, I'm tagging that one as ambiguous based on the data available from his BAM file - he has 5G reads for the derived allele and 2A reads for the ancestral allele at that position. Unfortunately, there aren't currently any phyloequivalent or downstream SNPs for I-FGC21819, so I'm guessing that this is the last point of commonality for the three folks in the databases of modern testers.

    I'll also suggest to YFull that they add SZ45 to their list of ancient DNA samples. Based on comparison with others I have suggested, I think SZ45 has a good chance of making it.

    Here's to a good 2019 for ancient I-M253 DNA (and modern descendants).

  2. The Following 10 Users Say Thank You to deadly77 For This Useful Post:

     Artmar (01-11-2019),  JMcB (01-11-2019),  JonikW (01-11-2019),  Megalophias (01-12-2019),  Nibelung (01-11-2019),  palamede (01-14-2019),  Pribislav (01-11-2019),  Sassoneg (01-19-2019),  slievenamon (01-15-2019),  Username (01-12-2019)

  3. #112
    Registered Users
    Posts
    955
    Sex
    Location
    Kent
    Ethnicity
    Isles Celto-Germanic
    Nationality
    British
    Y-DNA
    I1 Z140+ A21912+
    mtDNA
    V

    Wales England Cornwall Scotland Ireland
    That's quality detective work and much appreciated. I'd like to join you in wishing us a wonderful year for I1 discoveries. I hope our picture of ancient distribution patterns will be much improved by the year end.
    Living DNA Cautious mode:
    South Wales Border-related ancestry: 86.8%
    Cornwall: 8%
    Cumbria-related ancestry: 5.2%
    Y line: Peak District, England. Big Y match: Scania, Sweden; TMRCA 1,280 ybp (YFull);
    mtDNA: traces to Glamorgan, Wales, 18th century. Mother's Y line (Wales): R-L21 L371

  4. The Following 3 Users Say Thank You to JonikW For This Useful Post:

     deadly77 (01-11-2019),  JMcB (01-11-2019),  spruithean (01-11-2019)

  5. #113
    Registered Users
    Posts
    762
    Sex
    Location
    Ontario
    Nationality
    Canadian
    Y-DNA
    I-A14097

    Canada Netherlands United Kingdom Ireland France
    Wow! Great work, and here's to a productive year for ancient I1! I'm hoping we find some ancient Z140 or perhaps something upstream or downstream.
    Y-DNA: I-A14097(Scotland),
    Big Y: I-F2642>Y1966>Y3649>A13241>Y3647>A14097 (1,850 YBP)
    mtDNA: pending (Westeremden, Netherlands)
    Other lines:
    R-M222 x2, R-L21 x2, I-M223, R-S1141, R-U198 & R-U106, mtHg J1c3
    Known ancestry
    Paternal: Britain & Ireland, France and Germany
    Maternal: Netherlands

  6. The Following 3 Users Say Thank You to spruithean For This Useful Post:

     deadly77 (01-11-2019),  JMcB (01-11-2019),  JonikW (01-11-2019)

  7. #114
    Registered Users
    Posts
    379
    Sex
    Location
    United Kingdom
    Ethnicity
    European
    Nationality
    British
    Y-DNA
    I-L338
    mtDNA
    J1c1

    United Kingdom England England North of England Norfolk Scotland Ireland
    I also took a look at CL63 from the same paper. This one is from the other cemetery, Collegno in Northern Italy. This was assigned as I1a3, but with last downstream SNP as Z79. As we've discussed on previous posts on this thread, Z79 is in I2 rather than I1.

    Worth noting that unlike SZ45 (which was a WGS), for CL63 In-solution capture was performed, targeting ~1,237,207 SNPs (“1240K capture”) scattered across the nuclear genome. The enriched DNA was then sequenced on a NextSeq (75bp single-end run or 150 cycles paired-end run). So we shouldn't expect the coverage that we observed for SZ45, but worth a look to see what we can get.

    For Z79, that position lights up but there are 2G reads for derived and 1A read for ancestral. I'd call this ambiguous.

    Looking at the I-DF29 branch and phyloequivalent SNPs, also pretty ambiguous - DF29 itself is a single negative call, Z2891 is a single positive call, no calls for Z2890, Z2891, Z2893. Not really sure we can claim anything of significance here.

    Checked a few of the SNPs that they report for I1, namely M253 (3T), L81 (3C), L840 (2G) and they all check out.

    I checked Z63 for the I1a3 reported in Supplementary Table 3 - that lights up with 11 reads for C (derivative), so I'd say that this is one of the SNPs they were targetting. I tried some of the downstream SNPs under Z63 but didn't come up with much. No calls for BY151, BY3394, PH5281, Y5940, BY351, L849, S11545, Y8331, S2077, S2078, L1237, S10360, BY1933.

  8. The Following 4 Users Say Thank You to deadly77 For This Useful Post:

     JMcB (01-11-2019),  JonikW (01-11-2019),  slievenamon (01-15-2019),  Username (01-12-2019)

  9. #115
    Registered Users
    Posts
    955
    Sex
    Location
    Kent
    Ethnicity
    Isles Celto-Germanic
    Nationality
    British
    Y-DNA
    I1 Z140+ A21912+
    mtDNA
    V

    Wales England Cornwall Scotland Ireland
    Please keep up the fantastic work. It's great to have someone with the skills and tools to do this for I1 aDNA as I've seen so much good work on our R1b U106 fellow northerners. I hope you'll have a few more samples to look at before the year is out, with much exciting discussion here.
    Living DNA Cautious mode:
    South Wales Border-related ancestry: 86.8%
    Cornwall: 8%
    Cumbria-related ancestry: 5.2%
    Y line: Peak District, England. Big Y match: Scania, Sweden; TMRCA 1,280 ybp (YFull);
    mtDNA: traces to Glamorgan, Wales, 18th century. Mother's Y line (Wales): R-L21 L371

  10. The Following 2 Users Say Thank You to JonikW For This Useful Post:

     deadly77 (01-11-2019),  JMcB (01-12-2019)

  11. #116
    Registered Users
    Posts
    379
    Sex
    Location
    United Kingdom
    Ethnicity
    European
    Nationality
    British
    Y-DNA
    I-L338
    mtDNA
    J1c1

    United Kingdom England England North of England Norfolk Scotland Ireland
    Cheers folks. The tools are mostly free to access - the Broad Institute's IGV for viewing and indexing the BAM files; European Nucleotide Archive (ENA) and National Center for Biotechnology Information (NCBI) for hosting the raw data and making it available for download; phylogenetic trees: YFull tree and FTDNA public haplotree; looking up positions of SNPs on YBrowse, YFull database and Brad Larkin's Genetic Homeland site - the latter two are extremely helpful as they include positions on the hg19 reference as well as hg38 reference - the majority of these BAM files are in hg19 format. The YFull database requires a one-time fee for access but in my opinion very much worth the price but the rest are open access. The skills are largely self taught and involved a fair bit of amateur blundering along the way - proper practitioners of genetics and bioinformatics would likely be highly amused at some of the approaches I went down. So it's mostly down to interest and time availability. But it's been a good learning experience and I've had fun working through these. I appreciate the interest and encouragement from the comments.

    By my count, I've been through 13 of the 15 post-CE I1 individuals mentioned in this thread (6 Iceland, 3 Bavaria, 1 Sweden, 1 England, 1 Hungary, 1 Italy), some of which I've been able to assess in greater detail than some others, depending on the quality of the raw data. Should probably put together a summary in case there's a future avalanche of ancient I1...

    The two that I haven't got to yet are the two samples from Poland KO_55 and ME_7 reference from Zenczak 2017. Does anyone have a link to the paper? All I have been able to find is this link on Academia here https://www.academia.edu/33791135/20..._DNA_libraries - this appears to be an abstract to a presentation which lists eight I1 individuals including three I-L1237 and one I-Z59. Has anything with a bit more detail been published?

    I also tried searching for KO_55 and ME_7 at the ENA. I was unable to find any samples with these names, but I was able to find data for "Kow_55" here https://www.ebi.ac.uk/ena/data/view/SRS1815340 and "Mar_7" here https://www.ebi.ac.uk/ena/data/view/SRS1815413 - does anyone know if these are the correct ones?

    Unfortunately, there are no BAM files associated with either sample at ENA. They do have FASTQ files - I have tried downloading these and converting them to BAM files using Felix Immanuel's SRA/FASTQ to BAM kit. However, the resultant BAM file isn't readable using IGV so I'm assuming there was some error or corruption during the conversion. It looks from the website that the developer is focused on other pursuits and hasn't updated these tools for a while (and doesn't seem likely to do so in the future). Are there any alternatives for converting a FASTQ to a BAM? I have a laptop running Windows and no familiarity with Linux systems.

  12. The Following 4 Users Say Thank You to deadly77 For This Useful Post:

     JMcB (01-12-2019),  JonikW (01-12-2019),  Megalophias (01-12-2019),  Username (01-13-2019)

  13. #117
    Registered Users
    Posts
    256
    Sex
    Ethnicity
    Slavic
    Nationality
    Serb
    Y-DNA
    I2-PH908

    Serbia Montenegro Bosnia and Herzegovina Croatia Split-Dalmatia
    Quote Originally Posted by deadly77 View Post
    The two that I haven't got to yet are the two samples from Poland KO_55 and ME_7 reference from Zenczak 2017. Does anyone have a link to the paper? All I have been able to find is this link on Academia here https://www.academia.edu/33791135/20..._DNA_libraries - this appears to be an abstract to a presentation which lists eight I1 individuals including three I-L1237 and one I-Z59. Has anything with a bit more detail been published?

    I also tried searching for KO_55 and ME_7 at the ENA. I was unable to find any samples with these names, but I was able to find data for "Kow_55" here https://www.ebi.ac.uk/ena/data/view/SRS1815340 and "Mar_7" here https://www.ebi.ac.uk/ena/data/view/SRS1815413 - does anyone know if these are the correct ones?

    Unfortunately, there are no BAM files associated with either sample at ENA. They do have FASTQ files - I have tried downloading these and converting them to BAM files using Felix Immanuel's SRA/FASTQ to BAM kit. However, the resultant BAM file isn't readable using IGV so I'm assuming there was some error or corruption during the conversion. It looks from the website that the developer is focused on other pursuits and hasn't updated these tools for a while (and doesn't seem likely to do so in the future). Are there any alternatives for converting a FASTQ to a BAM? I have a laptop running Windows and no familiarity with Linux systems.
    The Zenczak study, with Y-DNA and autosomal data, is yet to be published, hopefully this year, since mt-DNA study has been published last year (Stolarek et al.). So I doubt BAM files are available. Thanks for looking all these samples more closely, maybe you could look Nordic LN/BA samples from Allentoft study next. They are poor quality, but shotgun sequenced nevertheless, so you could get lucky with some downstream SNPs. Here's a recap of what we have thus far:

    anc-i1.PNG b.PNG

  14. The Following 4 Users Say Thank You to Pribislav For This Useful Post:

     deadly77 (01-12-2019),  JMcB (01-12-2019),  JonikW (01-12-2019),  Username (01-13-2019)

  15. #118
    Registered Users
    Posts
    379
    Sex
    Location
    United Kingdom
    Ethnicity
    European
    Nationality
    British
    Y-DNA
    I-L338
    mtDNA
    J1c1

    United Kingdom England England North of England Norfolk Scotland Ireland
    Quote Originally Posted by Pribislav View Post
    The Zenczak study, with Y-DNA and autosomal data, is yet to be published, hopefully this year, since mt-DNA study has been published last year (Stolarek et al.). So I doubt BAM files are available. Thanks for looking all these samples more closely, maybe you could look Nordic LN/BA samples from Allentoft study next. They are poor quality, but shotgun sequenced nevertheless, so you could get lucky with some downstream SNPs. Here's a recap of what we have thus far:

    anc-i1.PNG b.PNG
    I've had a brief look at the four RISE samples from the Allentoft but not systematically worked through them. I hadn't written about them because to be honest, there's not a lot to report. You're correct that they are poor quality and as a result there's no read at most of the positions that I've looked at. All I've found so far are a really small bunch of one read negatives among some of the downstream SNPs. Most SNPs are no call, including all of the SNPs phyloequivalent to DF29. In fact among the >300 SNPs that define the I1 block, the four RISE samples are no call for at least 41 SNPs (including M253 itself) and I haven't had the motivation to go back and look at the rest yet. I'm not sure that they are fully I1 (as in, post the TMRCA of modern descendants) and may more appropriately be called pre-I1, proto-I1, I*, extinct lineage or whatever. Given the poor quality of the coverage in the BAM files it may be impossible to say one way or another. At least with the majority of the post-CE individuals I've been able to assign DF29 or one of the phyloequivalent SNPs so don't have to worry about holes upstream in the I1 block.

  16. The Following 5 Users Say Thank You to deadly77 For This Useful Post:

     JMcB (01-12-2019),  JonikW (01-12-2019),  Megalophias (01-12-2019),  Pribislav (01-12-2019),  Username (01-13-2019)

  17. #119
    Registered Users
    Posts
    490

    Quote Originally Posted by deadly77 View Post
    I've had a brief look at the four RISE samples from the Allentoft but not systematically worked through them. I hadn't written about them because to be honest, there's not a lot to report. You're correct that they are poor quality and as a result there's no read at most of the positions that I've looked at. All I've found so far are a really small bunch of one read negatives among some of the downstream SNPs. Most SNPs are no call, including all of the SNPs phyloequivalent to DF29. In fact among the >300 SNPs that define the I1 block, the four RISE samples are no call for at least 41 SNPs (including M253 itself) and I haven't had the motivation to go back and look at the rest yet. I'm not sure that they are fully I1 (as in, post the TMRCA of modern descendants) and may more appropriately be called pre-I1, proto-I1, I*, extinct lineage or whatever. Given the poor quality of the coverage in the BAM files it may be impossible to say one way or another. At least with the majority of the post-CE individuals I've been able to assign DF29 or one of the phyloequivalent SNPs so don't have to worry about holes upstream in the I1 block.
    Interesting that the earliest evidence of DF29+ is only from the Common Era.

    Considering that DF29 accounts for 99% of all I1 you would think that this SNP would have several reads for any ancient dna testing.

  18. #120
    Registered Users
    Posts
    379
    Sex
    Location
    United Kingdom
    Ethnicity
    European
    Nationality
    British
    Y-DNA
    I-L338
    mtDNA
    J1c1

    United Kingdom England England North of England Norfolk Scotland Ireland
    Quote Originally Posted by mwauthy View Post
    Interesting that the earliest evidence of DF29+ is only from the Common Era.

    Considering that DF29 accounts for 99% of all I1 you would think that this SNP would have several reads for any ancient dna testing.
    It's not that these pre-CE samples are DF29- (or the five phyloequivalent SNPs). It's that they are no call - there is no information to assess whether they are DF29+ or DF29-. In the BAM file at that position it's just blank, so it's a Schrodinger's cat situation. Some of the ancient DNA BAM files have pretty decent coverage - HSJ-A1 and SZ45 for example. Some have less where there's not a lot of information that you can reliably call from looking at the BAM file. The RISE samples are in that latter category. I looked at the RISE samples for a few more of the block of SNPs that make up I1 before it divides into descendant subclades. Out of the 70 SNPs looked at so far, there are only 4 SNPs read across the four samples, and they are all one-read. So that's 4/(4x70)=1.4% of just getting a read on SNPs looked at so far. And none of the four SNPs with a reading are showing up in more than one sample so far.

    I think one major factor is when the analysis was conducted. If you look at the publication dates, all but one of the post-CE samples I've looked are from papers published in 2018. The only one that wasn't was published in 2016. All of the pre-CE samples are from papers published in 2015 or 2014. Probably a bit of difference in the technology of sequencing. Also, the ancient remains have had a lot more time to degrade before they were analyzed.

    It may be that some of the RISE samples are not positive for some of the SNPs in the I1 block and may be an extinct lineage that doesn't have descendants that are alive today (or at least not in any database of modern testers that I'm aware of). In that case, wouldn't expect them to be DF29+. But if it's all no calls or one-read positives or negatives, it may be impossible to say with any certainty. We've seen above that CL63 has a single read negative call for DF29, which is clearly bogus due to a much stronger call for Z63 with 11 positive reads. And then SVK-A1 has a single false positive for Z131+ which saw SVK-A1 assigned as I1b in the paper, but when digging further into the data shows he's actually I-FGC21682 based on stronger SNP reads (and therefore must be Z131-). Isolated single read SNPs for derived or ancestral alleles may not be conclusive as we have examples of them turning up false, so we need to approach such calls with a bit of caution.

  19. The Following 4 Users Say Thank You to deadly77 For This Useful Post:

     JonikW (01-14-2019),  Michał (01-14-2019),  palamede (01-14-2019),  slievenamon (01-15-2019)

Page 12 of 23 FirstFirst ... 2101112131422 ... LastLast

Similar Threads

  1. Map of ancient E samples
    By Passa in forum E
    Replies: 73
    Last Post: Yesterday, 12:41 PM
  2. List of ancient samples on GEDmatch
    By Tomenable in forum Ancient (aDNA)
    Replies: 388
    Last Post: 03-05-2019, 10:57 PM
  3. Replies: 37
    Last Post: 05-07-2018, 12:05 AM
  4. My list of ancient European y-dna samples
    By venustas in forum Ancient (aDNA)
    Replies: 2
    Last Post: 10-14-2017, 05:19 AM
  5. Replies: 18
    Last Post: 05-01-2015, 11:26 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •