Y-Chromosome Adam and Noah

Xue, Yali et al.  “Human Y Chromosome Base-Substitution Mutation Rate Measured by Direct Sequencing in a Deep-Rooting Pedigree.”  Current Biology 19(17):1453–1457.  Sep 15, 2009.  A direct sequencing study that compared the DNA of two men from one Chinese family, whose common ancestor was born about 1805, and counted the mutations between them.  It gives a Y mutation rate of 1.0 × 10⁻⁹ changes per base per year, and it found that eight of the twelve candidate mutations had arisen in the cell lines rather than in the men.  The article uses it as its direct measurement of the Y rate, and notes that the rate rests on only four mutations.

1. 1a.

   Page 1456, section Discussion:  “No discrepancy in the pedigree or evolutionary rate was evident.”  The same paragraph shows the authors were speaking directly to the mitochondrial controversy: "this has been done for the mtDNA HVSI, where a controversy has emerged over whether the 'pedigree rate' measured in family studies is consistent with the 'evolutionary rate' inferred from comparisons of different species or whether it is substantially faster." ↩

2. 1b.

   Section Summary, page 1453:  “Twelve mutations were confirmed in ∼10.15 Mb; eight of these had occurred in vitro and four in vivo.  The latter could be placed in different positions on the pedigree and led to a mutation-rate measurement of 3.0 × 10−8 mutations/nucleotide/generation (95% CI: 8.9 × 10−9–7.0 × 10−8)”. ↩

3. 1c.

   Section Results, Mutation Rate, page 1455:  “The chromosomes were separated by 13 generations, and the common ancestor of the two individuals was born in approximately 1805.  The mutation rate is therefore 1.0 × 10−9 mutations/nucleotide/year (95% CI: 3.0 × 10−10–2.5 × 10−9), or 3.0 × 10−8 mutations/nucleotide/generation (95% CI: 8.9 × 10−9–7.0 × 10−8).”  Note for readers checking Jeanson & Holland against this: their Methods describe Xue's as a "two-person, 17-generation pedigree."  The paper says thirteen generations, in the abstract, the Results, the "Mutation Rate" section and the Experimental Procedures.  The error is harmless because they used Xue's published rate rather than recomputing it. ↩ ↩

4. 1d.

   Page 1456:  “We then tested all candidate mutations by capillary sequencing, and thus we are confident that the false-positive rate in the final dataset is effectively zero.”  “we can use this measurement to estimate a false-negative rate of <2% at the positions that differ between the DFNY1 and reference sequences.” ↩

Jeanson, Nathaniel T., and Ashley D. Holland.  “Evidence for a Human Y Chromosome Molecular Clock: Pedigree-Based Mutation Rates Suggest a 4,500-Year History for Human Paternal Inheritance.”  Answers Research Journal 12:393–404.  2019.  A genetics paper by two young-earth creationists, Nathaniel Jeanson of Answers in Genesis and Ashley Holland of Cedarville University.  It argues that the published pedigree studies of Y-chromosome mutation rates used low-coverage sequencing, which misses real mutations, and that high-coverage data give a rate that fits a 4,500-year history.  The article treats this as the young-earth case on the Y chromosome and tests its central causal claim against Helgason's published data.

1. 2a.

   Section Abstract:  “Pedigree-based mutation rates act as an independent test of the young-earth creation and evolutionary timescales.  Currently, evolutionists use published Y chromosome pedigree-based mutation rates to argue for an ancient origin of humanity.  However, their published studies rely on low-coverage sequence runs.  We show that pedigree-based mutation rates from high-coverage sequence runs are hidden in the evolutionary literature, and we demonstrate that these rates confirm a 4,500-year history for human paternal ancestry.” ↩

2. 2b.

   “the two high coverage datasets revealed a per-generation mutation rate that was, on average, 10- to 17-times faster (table 1) than the previously published (i.e., Xue et al. 2009; Helgason et al. 2015) low coverage studies.”  Their Table 1 lists sequence coverage of 15.5x for Xue, 12.4x for Helgason, 35.5x for Karmin and 40x for Maretty, with per-generation rates of 3.0 × 10⁻⁸, 3.00 × 10⁻⁸, 3.02 × 10⁻⁷ and 5.0 × 10⁻⁷.  The Helgason interval they print, 2.85 × 10⁻⁸ to 3.16 × 10⁻⁸, appears nowhere in Helgason's abstract; recomputing it from Helgason's Supplementary Table 9 by an exact Poisson interval gives 2.77 × 10⁻⁸ to 3.25 × 10⁻⁸, so theirs is a slightly narrower normal approximation of the right quantity.  The column head misspells Xue as "Xu." ↩ ↩

3. 2c.

   “From Karmin et al. (2015), we obtained the results of their high coverage sequencing runs on 24 Dutch father-son pairs, 6 Estonian father-son pairs, and 1 Estonian brother-brother pair from Table S2, column 'FS'.”  They derive Karmin's Y coverage as half the whole-genome figure:  “Because of the pairing of the Y chromosome with the X chromosome, the Y chromosome exists in an effectively haploid state.  Consequently, to calculate sequence coverage for the Y chromosome, we took the reported whole-genome sequencing coverage and divided it by 2.” ↩

4. 2d.

   On how the second rate was obtained:  “First, a screenshot was taken from the published pdf containing Figure 4c, and the screenshot included the associated scale bar of '50 mutations' (see Supplemental fig. 1).  (The raw sequence data for the Y chromosomes is restricted access.)  After electronically expanding the screenshot to a large size (while keeping the proportions constant), we located the 17 pairs with the shortest distances between them, to identify the 17 father-son pairs.  Then we used the scale bar to visually and electronically estimate the nucleotide differences between each father-son pair.” ↩

5. 2e.

   “it will be important to confirm these results with the raw (i.e., pairwise alignment-based rather than phylogenetic tree-based) Y chromosome sequences from the father-son pairs.  We anticipate little deviation from the results we described here.” ↩

6. 2f.

   On Karmin's Sanger validation:  “the Karmin et al. (2015) attempts to identify positive controls via Sanger sequencing add little of scientific value to our discussion.  They raise more questions than they answer.”  And their own reductio:  “Extrapolated to the rest of the father-son pairs, the Sanger sequencing results imply that all father-son differences were false positives, which would produce a mutation rate less than 1 mutation every 31 generations—10 times lower than expectations based on a model which puts the first male ancestor about 250,000 years ago.” ↩

7. 2g.

   In the conclusion:  “Our results demonstrate that a Y chromosome molecular clock exists, and that it specifies about 4,500 years in total for human paternal history (figs. 1, 3).”  And the concession inside it:  “these predictions captured all but the A00 branch length values … These latter results predicted a higher per-generation mutation rate for the most divergent A00 individuals.” ↩

Carter, Robert W.  “Can we place the sons of Noah on the Y chromosome tree?”  Creation Ministries International.  Oct 29, 2024.  A short article by Robert Carter, a creation geneticist, on whether the three sons of Noah can be placed on the human male family tree.  He concludes that nine different hypotheses fit the Bible's constraints, so the shape of the Y tree cannot show which lineages came from Noah's sons.  The article uses it to show that Carter himself lists nine compatible hypotheses, which permit any branching pattern.

1. 3a.

   “Each of these nine hypotheses comport to the Scriptural constraints.  The three Y chromosomes of Noah's sons could have been identical or different.  They could have been slightly different or highly divergent.  There could have been one, two, or three different versions, and these versions could form a nested relationship, or not.  Did I miss anything?”  This is the most intellectually honest document on the young-earth side of this question.  Its other admissions, in order:  “If there was little mutation prior to the Flood, the three sons of Noah could have had identical Y chromosomes … There would not be 'three' founding branches, there would only be 'one'”.  “there is no real way to really know the sequence of 'Y chromosome Noah'”.  “We cannot know, therefore, that the rate of mutation accumulation was the same in every post-Flood family.  Thus, we cannot know if the different branch lengths near the middle of the tree represent 'time' or 'random chance'”.  And the warning to his own side:  “we need to be very careful about making firm conclusions when too many alternate hypotheses exist.” ↩ ↩

Helgason, Agnar et al.  “The Y-chromosome point mutation rate in humans.”  Nature Genetics 47(5):453–457.  Mar 25, 2015.  A genetic study of 753 Icelandic men grouped into 274 family lines, which counted new point mutations (single-letter changes in DNA) on the Y chromosome across 1,365 father-to-son transmissions, called meioses, covering 47,123 years.  It found 8.71 × 10⁻¹⁰ mutations per position per year for 15.2 of the 21.3 Mb of male-specific Y sequence, and a slightly lower rate on the palindromic (mirror-image) parts.  The article uses it as the largest direct measurement of the Y mutation rate.

1. 4a.

   Section Abstract:  “Here we used whole-genome sequence data from 753 Icelandic males, grouped into 274 patrilines, to estimate the point mutation rate for 21.3 Mb of male-specific Y chromosome (MSY) sequence, on the basis of 1,365 meioses (47,123 years).  The combined mutation rate for 15.2 Mb of X-degenerate (XDG), X-transposed (XTR) and ampliconic excluding palindromes (rAMP) sequence was 8.71 × 10(-10) mutations per position per year (PPPY).  We observed a lower rate (P = 0.04) of 7.37 × 10(-10) PPPY for 6.1 Mb of sequence from palindromes (PAL), which was not statistically different from the rate of 7.2 × 10(-10) PPPY for paternally transmitted autosomes.” ↩ ↩ ↩

2. 4b.

   Table Supplementary Table 6, supplement page 9:  “A subset of 101 candidate mutations were selected randomly for validation by means of Sanger sequencing, with a 10% bias for mutations on branches with only one descendant.”  “Out of the remaining 75 informative assays, 72 confirmed the presence of a mutation on the Y chromosome, whereas three did not.” ↩

3. 4c.

   Table Supplementary Table 6, supplement page 10:  “Based on the Sanger sequencing, we therefore report a false positive rate of 1/75 or 0.013 (95% C.I. 0.0007 - 0.08) for all branches.  However, when restricted to only the 482 branches with an average XDG sequence depth >10X, that were used in this study, the false positive rate becomes 0/43 (95% C.I. 0 - 0.102).” ↩

4. 4d.

   Table Supplementary Table 7, supplement page 11:  “In all cases, the reads mapped to the relevant positions in the three additional males confirmed the genotypes called in the original set of males from those patrilines.  The genotype error rate from this experiment is estimated to be 0/41 (95% C.I. 0-0.107) and the false-positive rate for calling mutations is estimated to be 0/15 (95% C.I. 0-0.253).” ↩

5. 4e.

   Table Supplementary Table 8, supplement page 12:  “Out of the 8956 genotypes that were valid from both platforms, two exhibited a mismatch, corresponding to an error rate of 0.00022 (95% C.I. 0-0.0009).”  Helgason also tested rate constancy across haplogroups directly, in the Supplementary Table 9 note: "Four of the 27 sequence regions yielded p-values <0.05 … In all four cases, the difference is due to an apparently higher mutation rate for haplogroup E1b1, which is represented by only five branches from a single patriline.  After adjustment for the number of tests, only one region (XDG9) remained significant."  The table's own total row gives a chi-square of 10.11, p = 0.072, for equality of the six haplogroup-wide rates. ↩

6. 4f.

   Table Supplementary Table 8, supplement page 12:  “We note that these two mismatches would not lead to a false positive call of mutation in our study design – which requires the mutant allele to be present in at least two copies on each strand.  Moreover, they are unlikely to give rise to false negatives, as pseudo-heterozygote genotypes were considered in our study design to detect de novo mutations.”  The only place in the paper that addresses false negatives, and it is an argument from the calling rule rather than a measurement. ↩

Poznik, G. David et al.  “Sequencing Y Chromosomes Resolves Discrepancy in Time to Common Ancestor of Males Versus Females.”  Science 341(6145):562–565.  Aug 2, 2013.  A Science report that estimated the common-ancestor age of all Y chromosomes and of all mitochondrial DNA, and found the Y ancestor at 138 ky (120–156 ky) and the mitochondrial one at 124 ky (99–148 ky).  It concluded that male lineages do not coalesce significantly more recently than female lineages.  The article uses it for the close Y and mitochondrial dates, and for a Y mutation rate of 0.82 × 10⁻⁹ calibrated on the entry of people into the Americas.

1. 5a.

   “We estimate the Y chromosome TMRCA to be 138 ky (120–156 ky) and the mtDNA TMRCA to be 124 ky (99–148 ky).”  Section Abstract:  “Our findings suggest that, contrary to prior claims, male lineages do not coalesce significantly more recently than female lineages.”  The title's claim that the discrepancy is "resolved" omits A00, published the same year, which reopens it in the other direction. ↩

2. 5b.

   “We estimate the ratio of the mtDNA TMRCA to that of the Y chromosome to be 0.90 (95% CI: 0.68–1.11; Fig. S13).  If, as argued above, the divergence of the Y chromosome Q lineages occurred at approximately the same time as that of the mtDNA A2 lineages, then the TMRCA ratio is invariant to the specific calibration time used.” ↩

3. 5c.

   “Consequently, the divergence between these two lineages provides an appropriate calibration point for the Y mutation rate.  The large number of variants that have accumulated since divergence, 120 and 126, contrasts with the pedigree-based estimate of the Y chromosome mutation rate, which is based on just 4 mutations.  Using entry to the Americas as a calibration point, we estimate a mutation rate of 0.82 × 10−9 /bp/yr (95% CI: 0.72–0.92 × 10−9 /bp/yr).”  And, independently, in the same paragraph:  “we consider the Out of Africa expansion of modern humans, which dates to approximately 50 kya and yields a similar mutation rate of 0.79 × 10−9 /bp/yr.” ↩ ↩

Karmin, Monika et al.  “A recent bottleneck of Y chromosome diversity coincides with a global change in culture.”  Genome Research 25(4):459–466.  Apr 2015.  A study of Y-chromosome sequences from 475 male samples that dates the common ancestor of all Y lineages in Africa to about 254,000 years, using ancient DNA samples to calibrate the mutation rate.  It also reports a second strong bottleneck in male lineages within the last 10,000 years, which the authors attribute to cultural change.  The article uses it for the Holocene male bottleneck and for the filter counts in its supplementary Table S2.

1. 6a.

   Section Abstract:  “Applying ancient DNA calibration, we date the Y-chromosomal most recent common ancestor (MRCA) in Africa at 254 (95% CI 192–307) kya and detect a cluster of major non-African founder haplogroups in a narrow time interval at 47–52 kya … In contrast to demographic reconstructions based on mtDNA, we infer a second strong bottleneck in Y-chromosome lineages dating to the last 10 ky.  We hypothesize that this bottleneck is caused by cultural changes affecting variance of reproductive success among males.” ↩

2. 6b.

   “we calibrated the Chr Y mutation rate in our CG data by using inferences of the coalescent times of two Chr Y haplogroups, Q1 and Q2b, from ancient DNA data.  We used Chr Y data of the 12.6-ky-old Anzick (Q1b) and 4-ky-old Saqqaq (Q2b) specimens … yielding the final estimate of 0.74 × 10−9 (95% CI 0.63–0.95 × 10−9) per bp per yr.”  And their own candid limitation in the same paragraph:  “It should be noted that this estimate is based on only two ancient DNA samples from a relatively recent time horizon and the same Y chromosome haplogroup.  However, a very similar mutation rate estimate of 0.76 × 10−9 per bp per yr was determined independently from a different ancient DNA specimen of much older age by a recent study (Fu et al. 2014).” ↩

3. 6c.

   “the Y chromosome plot suggested a reduction at around 8–4 kya (Supplemental Fig. S4B; Supplemental Table S4) when the female Ne is up to 17-fold higher than the male Ne.” ↩ ↩

4. 6d.

   “both mtDNA and Y plots (Supplemental Fig. S4) showed an increase of Ne in the Holocene, which has been documented before for the female Ne.” ↩

5. 6e.

   Section Supplementary Table S2:  The table is titled "The effect of different quality filters on the number of Y chromosome SNPs in 475 male samples," in the Supplemental Tables PDF.  Its own column definition:  “FS - average number of father – son differences in 24 Dutch father-son pairs, 6 Estonian father-son pairs and 1 brother-brother pair, including cases where father has an ancestral allele and the son or one of the brothers has a derived allele.”  The rows used in the filter-ladder table above are read directly from it: no filters, 82,306 SNPs at FS 180.26; filter c (regional exclusion mask), 10,793,302 callable bp, 42,340 SNPs, FS 3.26; and filter a+b+d, the combination the paper uses, 8,819,704 bp, 35,672 SNPs, FS 0.61.  Jeanson and Holland's headline "high coverage" rate is arithmetically exact: 3.26 / 10,793,302 = 3.02 × 10⁻⁷ per bp per generation, the filter-c row.  Nothing in their reading of this table is fabricated.  The table prints L = NA for the "<5% no-calls" row, so the ladder reuses the unfiltered 59,373,566 bp for it, which makes that row's implied age an upper bound. ↩ ↩

6. 6f.

   Section Supplementary Table S2:  “L - total length of chrY considered, including variable and non-variable positions; SNPs - total number of binary SNPs;”.  Both columns are printed beside the father-son column for every filter, for the same set of men. ↩

7. 6g.

   Section Supplemental Text, supplement page 9:  “The Saqqaq sequence shared the highest number of derived alleles in our data with four Koryak sequences in haplogroup Q2b-B143 (Figure S35).”  Supplement page 10:  “Considering that the four Koryak sequences are on average at a distance of 19.5 transversions from the most recent common ancestor they share with Saqqaq, we estimate that the Saqqaq misses 11.6 transversions per 4,000 years, which considering 1.7 as the average transition/transversion ratio in our a+b+d filtered data for 8,819,704 positions, yields a rate estimate of binary SNPs of 0.89x10-9 /bp/year.”  Karmin's nomenclature calls Saqqaq's subgroup Q2b;  Rasmussen's 2010 nomenclature calls it Q1a.  Both place it inside haplogroup Q. ↩ ↩

8. 6h.

   Section Supplemental Text, supplement page 7:  “For these six SNV-s we were able to design Chr Y specific primers and to get homozygous Chr Y sequence calls for five of these positions.  Sanger sequencing revealed that all of the studied positions were false positives.” ↩

9. 6i.

   Section Supplemental Text, supplement page 4:  “The number of FS [Father-Son] differences was approximately 10 fold higher than the expected number of de novo mutations considering the range of published Chr Y mutation rates … This finding prompted us to explore additional filters.” ↩

10. 6j.

    Section Supplemental Text, supplement page 4:  The sentence before it:  “We initially applied a combination of regional filters previously defined on the basis of analyses of Illumina HiSeq data (Poznik et al. 2013; Wei et al. 2013a), resulting in ten regions of Chr Y sequence, altogether capturing 10.8 Mb (filter c, Table S2).  However, the application of the regional filters led only to a modest reduction of false positive calls judged by the number of father-son/brother-brother (FS) differences and the count of recurrent mutations (Table S2).”  This fixes the referent: the ten-fold excess is Karmin's verdict on filter c, which is the row Jeanson and Holland take as their high-coverage rate. ↩

Petr, Martin et al.  “The evolutionary history of Neandertal and Denisovan Y chromosomes.”  bioRxiv.  Mar 9, 2020.  A genetics study that sequenced the first Denisovan Y chromosomes and three late Neandertal Y chromosomes, in a 104-page preprint with its supplementary information that had not been peer reviewed.  The peer-reviewed version appeared as Science 369(6511):1653–1656 on 25 September 2020, doi:10.1126/science.abb6460.  It dates the Denisovan Y split to about 700 thousand years ago and the Neandertal and modern human Y split to about 370 thousand years ago, and it argues that gene flow replaced the Neandertal Y and mitochondrial DNA.  The article uses it for the paper's estimate that the modern human Y root is about 249 kya, made with the paper's own mutation rate.

1. 7a.

   Page 6:  Preprint page 6.  This is the preprint's wording, not the published Science version, whose wording and figures may have changed in review.  “We first calculated the mutation rate in the total 6.9 Mb target regions to be 7.34×10-10 per bp per year (bootstrap CI: 6.27-8.46×10-10) … Using this mutation rate, we estimated the TMRCA of the African A00 lineage and a set of non-African Y chromosomes from the SGDP panel at ~249 kya (bootstrap CI: 213-293 kya) … These estimates are consistent with values inferred from larger-scale studies of present-day human Y chromosomes, suggesting that the Y chromosomal regions we defined for capture are not unusual in terms of their mutation rate.” ↩ ↩ ↩

2. 7b.

   The published abstract, Science 369(6511):1653, says:  “The phylogenetic relationships of archaic and modern human Y chromosomes differ from the population relationships inferred from the autosomal genomes and mirror mitochondrial DNA phylogenies, indicating replacement of both the mitochondrial and Y chromosomal gene pools in late Neanderthals.  This replacement is plausible if the low effective population size of Neanderthals resulted in an increased genetic load in Neanderthals relative to modern humans.”  The preprint abstract, on page 2, was stronger against itself:  Page 2:  “Although unlikely under neutrality, we show that this replacement is plausible if the low effective population size of Neandertals resulted in an increased genetic load in their Y chromosomes and mitochondrial DNA relative to modern humans.”  The clause Although unlikely under neutrality does not appear in the published abstract, so the concession the preprint made did not survive peer review and is reported here as preprint wording only. ↩

Poznik, G. David et al.  “Punctuated bursts in human male demography inferred from 1,244 worldwide Y-chromosome sequences.”  Nature Genetics 48(6):593–599.  Apr 25, 2016.  A study that sequenced 1,244 Y chromosomes from men around the world and built a dated family tree of the male lines, with a supplementary note and data archive giving the split times.  It reports bursts of male population growth on each of five continents, at times of known migrations and technological changes.  The article uses its backbone of split times, which places the root of the Y tree about 163,000 years ago when dated by the pedigree rate, and its statement that one split is a short internal branch.

1. 8a.

   Section Supplementary Note, tables Supplementary Table 10 (Split-time estimates), supplement page 47:  “T, estimated split time, assuming the aDNA-based mutation rate estimate of Fu et al. (μ = 0.76 × 10–9 per bp per year);  T′, estimated split time using the pedigree-based mutation rate estimate of Helgason et al. (μ = 0.888 × 10–9 per bp per year).”  The backbone, with T and T′ in thousands of years: root A0–T splits A0 from A1 at 190.4 / 163.0; A1 splits A1a from BT at 159.0 / 136.1; BT splits B from CT at 105.8 / 90.5; CT splits DE from CF at 76.0 / 65.0; CF splits C from F at 75.5 / 64.6.  The paper gives no confidence interval for the root of about 190,000 years anywhere in the text or supplement, and none has been invented here.  Poznik uses 0.888 × 10⁻⁹ as "the Helgason rate" rather than the 8.71 × 10⁻¹⁰ of Helgason's abstract, because Helgason's rate is region-specific and Poznik reweighted it to their own 10.3 Mb callable set. ↩ ↩ ↩ ↩

2. 8b.

   Section Supplementary Note 1.3.2, False Negative Rate, supplement page 59:  “We observed 17,194 sites in the CG data. Of these, 13,360 were called in both datasets and 3,834 (22%) were not called in the low-coverage data. Most of these (3,343; 87%) appear to be false negative singletons (Supplementary Figure 1).”  CG is Complete Genomics high-coverage sequence of 143 of the same men.  After four outlier samples and 535 sites near other variants were set aside, the share of unobserved sites was 14.2%. ↩

3. 8c.

   Section Supplementary Note, supplement page 81:  “The parent of hgC and megahaplogroup F, CF-P143, is another short internal branch.  These clades share just four SNPs, including the previously known P143.” ↩ ↩

4. 8d.

   Section Abstract:  “Our phylogeny reveals bursts of extreme expansions in male numbers that have occurred independently among each of the five continental superpopulations examined, at times of known migrations and technological innovations.”  The paper is also candid about calibration:  “For later male-specific expansions, we should consider the consequences of alternative mutation rate estimates, as pedigree-based methods relying on variation from the most recent several centuries may be more relevant.  The pedigree-based estimate from the largest set of mutations would lead to a decrease in expansion times by ~15%.” ↩

Porubsky, David et al.  “Human de novo mutation rates from a four-generation pedigree reference.”  Nature 643(8071):427–436.  2025.  A genome study of one four-generation, 28-member family, CEPH 1463, read with five short-read and long-read sequencing technologies and assembled almost end to end.  It counts new mutations of every kind on each transmission, and on the Y calls them from assembled chromosomes on five father-to-son transmissions, against the grandfather's own Y.  It is the most complete father-son Y measurement yet made, the test the young-earth coverage argument asks for, though it is one family with few mutations outside the satellite repeats. ↩

1. 9a.

   Page 427, section Abstract:  “Among male individuals, we find 12.4 de novo Y chromosome events per generation.”  The count includes small insertions, deletions and structural changes as well as single-base mutations, across the whole male-specific Y. ↩

2. 9b.

   Page 432, section Y chromosome mutations:  “In total, we identify 48 de novo SNVs in the MSY across the 5 G2–G3 male individuals, ranging from 7 to 11 SNVs per Y transmission (mean, 9.6; median, 10) (Supplementary Table 13).  Only 2 SNVs map to the Y euchromatic regions, 1 to the pericentromeric regions and the remaining 45 out of 48 map to the Yq12 heterochromatic satellite regions (Fig. 5b).  In total, we estimate a de novo SNV rate of 1.99 × 10−7 mutations per bp per generation (95% CI = 1.59 × 10−7–2.39 × 10−7) for the entire MSY.  This estimate is an order of magnitude higher than that previously reported for Y euchromatic regions … due to access to Yq12 satellite DNA (Supplementary Table 13).  We note that 13 out of 45 (29%) of the DNMs had 100% identical matches elsewhere in the Yq12 region (but not at orthologous positions) and probably result from interlocus gene conversion events within the DYZ1/DYZ2 repeats … (Methods).”  “Overall, 82% (51 out of 62) of the DNMs identified on chromosome Y (42 out of 48 SNVs; 4 out of 9 of indels; and 5 out of 5 SVs) are located in regions where short reads cannot be reliably mapped (mapping quality = 0).”  MSY is the male-specific region of the Y.  The two ellipses drop reference numbers;  the earlier report is reference 40, Helgason 2015. ↩ ↩ ↩ ↩

3. 9c.

   Page 430, figure Fig. 2 legend:  “A significant difference from the autosomal DNM or PZM rate was determined using two-sided t-tests; … 0.59 (assembly-based DNMs in chromosome Y euchromatic regions)”.  0.59 is the P value for the euchromatic Y rate against the autosomal rate.  DNM is a de novo mutation, PZM a postzygotic one. ↩

4. 9d.

   Section Methods, Calling de novo TRs:  “their G2 parents (NA12877 and NA12878) were sequenced to 99 and 109 HiFi sequencing depths, resulting in a far lower chance of parental allelic dropout than samples with more modest sequencing depths.”  NA12877 is the father of the four G3 sons whose Y chromosomes were compared with his.  HiFi depth is counted over both copies of the autosomes, so a single-copy chromosome such as the Y gets about half. ↩

Jeanson, Nathaniel T.  “Testing the Predictions of the Young-Earth Y Chromosome Molecular Clock: Population Growth Curves Confirm the Recent Origin of Human Y Chromosome Differences.”  Answers Research Journal 12:405–423.  2019.  A follow-up paper by Nathaniel Jeanson of Answers in Genesis that tests whether recent population growth shows up in the branches of the Y tree, using the primary data from Karmin et al. (2015).  It reports that this growth fits the 4,500-year timescale of his earlier paper, and that the same finding revises the root of the Y tree.  The article uses it to show that Jeanson's headline figure is computed at his Alpha root, not at the Gamma root where the Flood's clearest fingerprint would sit.

1. 10a.

   “The first criterion attempted to optimize the detection of triad-like structures near the root, as well as ethnically diverse lineages near the root.  The pursuit of triad-like structures arose from the hypothesis that all three sons of Noah—Shem, Ham, and Japheth—would have left male descendants that would be easily identified today by a small but global sampling of male Y chromosome lineages.”  The confirming tree he uses here is not built in this paper.  He wrote:  “I used my derived neighbor-joining tree (see Supplemental fig. 3 of Jeanson and Holland 2019) as a secondary confirmation of the findings from the published tree.”  Its construction is described in the companion paper, which drew "a midpoint-rooted neighbor-joining tree … with MEGA7 … selecting the 'Pairwise deletion' option" and took branch lengths "extracted manually from the tree" (Jeanson & Holland 2019, Materials and Methods).  Midpoint rooting places the root at the centre of the longest tip-to-tip path by definition, so it cannot itself be evidence about where the root is.  In fairness, his primary analysis runs on Karmin's published tree, from Figure S3 and Table S7, so the neighbour-joining figure is a check rather than the sole basis. ↩

2. 10b.

   “I observed that the Y chromosome tree possessed a type of triad structure near the junction of the haplogroup C lineages with the haplogroup A/B/D/E lineages and with the haplogroup F-through-T lineages (i.e., node 5 in Figure S3 of Karmin et al. 2015).  This structure did not appear to be an artifact of the Karmin et al. (2015) analysis as it was also present in the 1000 Genomes Project data (see Supplementary Figure 14 of Poznik et al. [2016], especially node 72 on page 16 of the Supplementary Information).”  Branch 72 is the CF-P143 branch in Poznik's 1000 Genomes tree, so his identification of the feature is correct. ↩ ↩

3. 10c.

   “These two types of criteria were largely mutually exclusive because of their approaches to the math of population growth.”  The population-growth criterion.  “would produce a tree orientation that put the fewest number of lineages near the root … Thus, almost by definition, this mathematical optimization method avoided the pursuit of triad-like structures.”  The Alpha root he settles on is described as being.  “near the center of the unbranching line that connected the junction of haplogroups A through E, to the junction of haplogroups F through T.” ↩ ↩

4. 10d.

   The scores for how much of global male population growth each candidate root captures:  “the evolutionary root "captured only 27%"; a hypothetical Neanderthal root "captured only 14%"; Gamma "About 90%"; Epsilon "About 94%"; Alpha "About 95%."”  He also links the root choice to patriarchal drive:  “because of Noah's advanced age (500 years) at the begetting of his sons, he may have contributed an unusual number of mutations to his 3 boys (e.g., see Carter 2019; Carter, Lee, and Sanford 2018).” ↩

Zeng, Tian Chen et al.  “Cultural hitchhiking and competition between patrilineal kin groups explain the post-Neolithic Y-chromosome bottleneck.”  Nature Communications 9:2077.  May 25, 2018.  A theory paper that brings together anthropological theory, earlier population-genomic studies and mathematical models.  It proposes that competition between patrilineal kin groups, groups whose members trace descent through their fathers, could cut male Y-chromosome lines so sharply that it explains the male-only bottleneck Karmin and colleagues inferred.  The article cites it as the mainstream explanation of that bottleneck, the one it argues against.

1. 11a.

   Section Abstract:  “An abrupt population bottleneck specific to human males has been inferred across several Old World (Africa, Europe, Asia) populations 5000–7000 BP.  Here, bringing together anthropological theory, recent population genomic studies and mathematical models, we propose a sociocultural hypothesis, involving the formation of patrilineal kin groups and intergroup competition among these groups.  Our analysis shows that this sociocultural hypothesis can explain the inference of a population bottleneck.” ↩ ↩

2. 11b.

   Section Introduction:  “Karmin et al. inferred an intense bottleneck in Y-chromosomes in various geographical regions of the Old World around 5000–7000 BP (Fig. 1), suggesting a decline in the male effective population size during the Neolithic to approximately one-twentieth of its original level before the Neolithic in regions including Africa, Europe, Asia and the Middle East.”  This is a decline relative to the same population's earlier male size, not a male-to-female ratio, and Zeng is restating Karmin rather than measuring it independently.  Zeng also calls Karmin's "1:17 disparity" one that.  “falls far outside ratios typical in ethnographic data.” ↩

Mendez, Fernando L. et al.  “The Divergence of Neandertal and Modern Human Y Chromosomes.”  American Journal of Human Genetics 98(4):728–734.  Apr 7, 2016.  A genetics study that sequenced about 120 kb of Y chromosome DNA from one Neandertal man from El Sidrón, Spain, and compared it with chimpanzee and modern human Y chromosomes.  It estimated that the Neandertal and modern human Y lines split about 588 thousand years ago, and it says the Neandertal Y lineage is most likely extinct.  The article uses it as a second estimate of the modern human Y root, 275 kya, calibrated on a radiocarbon-dated genome.

1. 12a.

   “we estimated that T_AR = 275 kya (95% CI: 241–305 kya)”.  This paper's headline 588 kya Neandertal-modern split was later shown by Petr et al. to have been inflated by permissive filtering of a low-coverage archaic branch.  The authors themselves warned that the Neandertal-specific branch.  “is enriched for false positives as a result of low coverage, DNA damage, and sequencing errors.” ↩ ↩

Rasmussen, Morten et al.  “Ancient human genome sequence of an extinct Palaeo-Eskimo.”  Nature 463(7282):757–762.  Feb 11, 2010.  A genome study that sequenced DNA from a tuft of hair of one man, the Saqqaq individual, a member of an extinct Palaeo-Eskimo people.  The hair was dated by radiocarbon, a method that measures how much of the carbon-14 isotope a sample has left.  The article uses it because the date falls inside the window that young-earth writers accept, and because his Y chromosome belongs to haplogroup Q1a, a named branch of the Y family tree.

1. 13a.

   Section Results:  “Accelerator mass spectrometry (AMS) radiocarbon dating of the hair sample produced a date of 4,044 ± 31 14C yr bp and 4,170–3,600 cal. yr bp when correcting for local marine reservoir effect.” ↩ ↩

2. 13b.

   “our analysis of more than two hundred Y chromosome SNPs … allowed us to assign the Saqqaq individual to Y chromosome haplogroup Q1a …, commonly found among Siberian and Native American populations.”  Two parenthetical cross-references, "(Supplementary Information)" and "(Fig. 3d)", are elided at the ellipses. ↩ ↩

Mendez, Fernando L. et al.  “An African American Paternal Lineage Adds an Extremely Ancient Root to the Human Y Chromosome Phylogenetic Tree.”  American Journal of Human Genetics 92(3):454–459.  Mar 7, 2013.  A short report in which a team led by Fernando Mendez sequenced the Y chromosome of an African American man whose lineage, called A00, became a new and very old root of the Y tree.  The authors date that root to about 338 thousand years, older than the oldest anatomically modern human fossils and older than their own estimate for mitochondrial DNA.  The article uses it for that date, and for the point that one previously unsampled lineage can move a single-locus date by roughly 200,000 years.

1. 14a.

   Section Summary:  “We then estimated the time to the most recent common ancestor (TMRCA) for the Y tree as 338 thousand years ago (kya) (95% confidence interval = 237–581 kya).  Remarkably, this exceeds current estimates of the mtDNA TMRCA, as well as those of the age of the oldest anatomically modern human fossils.”  The 338 kya figure is disputed inside the mainstream.  Elhaik, Tatarinova, Klyosov and Graur re-estimate it at 209,500 years (95% CI 168,000–257,400) from the raw counts of 43 differences in 240,000 bp on A00 and 45 in 180,000 bp on A0, and quote Thomas Krahn, who found A00, saying by permission that "Fernando [Mendez] insisted using autosomal SNP frequency data for dating, which I thought is absolute nonsense for the Y mutation mechanism."  Every figure in that dispute is between 200,000 and 580,000 years.  The authors printed their mutation-rate range an order of magnitude wrong, and the journal corrected it in AJHG 92:632. ↩

2. 14b.

   “If we were to use the higher mutation rate (1.0 × 10−9 per base per year) rather than a realistic range derived from whole-genome sequencing … the estimated TMRCA for the tree incorporating A00 as the basal lineage would be 209 kya … We note, however, that the higher mutation rate produces an estimate for the common ancestor of all non-African Y chromosome haplogroups (C through T) of ∼39 kya … It is difficult to reconcile the younger estimate with the timing of the out-of-Africa dispersal on the basis of the analyses of autosomal DNA and the fossil record outside of Africa.” ↩

3. 14c.

   “This underscores how the stochastic nature of the genealogical process can affect inference from a single locus.” ↩ ↩

4. 14d.

   Paragraph 2:  The Y rate is not measured on the Y but transferred from autosomal trios.  “we developed a likelihood-based method that uses mutation rates recently estimated by Kong et al.,14 who performed high-coverage whole-genome sequencing on 78 human pedigrees.”  Paragraph after Figure 1:  The divisor is given in the paragraph immediately after Figure 1:  “b ∼ N(2.01 [SD = 0.17]) … the linear coefficient in the linear fit of mutations as a function of paternal age (2.01 per year [SEM = 0.17 per year]) … We then divided those values by 2.68 × 10⁹ (the length of autosomal sequence covered in each of the 78 trios).”  The 2.68 Gb divisor is Mendez's; Kong 2012 itself prints 2.63 billion. ↩

Jeanson, Nathaniel T.  “Response to: The Promises and Pitfalls of Correlating Y Chromosome Genetics to Human History, Review of Traced by Nathaniel T. Jeanson.”  Answers Research Journal 18:153–168.  Mar 26, 2025.  Jeanson's reply to Robert Carter's 2022 review of his book Traced, the latest statement of the young-earth Y clock by its author.  It restates the fast-rate argument in his own words, quotes Traced at length, explains why he leaves ancient DNA out of his analyses, dates the arrival of Native American Y lineages, and claims that newer data confirm his predictions.

1. 15a.

   Section Pre-Traced: Early Experiments:  “We found that high quality (high coverage) studies produced fast mutation rates—about three mutations per generation. … Low quality (low coverage) produced slow mutation rates—as expected from the known data on the relationship between sequencing coverage and variant discovery (for example, see Supplementary Figure 1 from Poznik et al. 2016 and discussion surrounding it).” ↩

2. 15b.

   Section Pre-Traced: Early Experiments:  “Surprisingly, the evolutionary community explicitly or implicitly filtered out results that disagreed with their expectations (see Jeanson and Holland 2019 for documentation and details).” ↩

3. 15c.

   Section “Finding Noah”:  “Recall that, on average, three Y chromosome mutations represent a single generation.  Up to five mutations might as well.  But three is the number common between two independent studies, and it is precisely the number needed to explain the length of the branches on the Y chromosome tree.”  Jeanson quoting chapter 13 of his book Traced (2022), pages 176–180 by his own citation. ↩

4. 15d.

   Section Pre-Traced: Tests on Ancient DNA:  “If I included Neanderthal DNA, the reconstruction captured only 14% (Jeanson 2019). … From this point on in my research, I omitted ancient DNA from further analysis.” ↩

5. 15e.

   Section “Molecular Clocks Create Highly Unlikely Scenarios”:  “The timing is more than just a few centuries; it’s roughly 1,500 years ago (that is, the A.D. 300s to 600s).  Also, my model predicts that a more ancient lineage might still be present among Native Americans, albeit at very low levels.”  Answering Carter's statement that the model forces a complete replacement of Native American Y chromosomes a few centuries ago. ↩

6. 15f.

   Section “Molecular Clocks Create Highly Unlikely Scenarios”:  “In short, new branches are still being discovered—all throughout the tree, including at the tips of the tree, in the middle sections, and in its deepest parts. … In contrast, the FTDNA haplogroup F-F15527 contains four deep F branches, all of which connect to living men.”  FTDNA is FamilyTreeDNA, a commercial testing company whose database Jeanson describes as about 200,000 men, mostly of European descent. ↩

Computed for this article in research/code/porubsky-y-test.ts from Porubsky's Supplementary Table 13.  Accessible X-degenerate bases summed over the five sons are 43,017,616, and all euchromatic bases 110,662,556.  Both euchromatic single-base mutations fall in X-degenerate sequence, one in NA12877 and one in NA12886.  An expected count is the rate times the summed bases, and each chance is the Poisson probability of 2 or fewer.  Helgason's per-generation rate is his X-degenerate 8.884 × 10⁻¹⁰ per year times his 34.5-year mean generation.  The upper bounds are exact Poisson 95% limits on 2 events, 7.22.  The comparison is per generation and does not adjust for the fathers' ages in this family. ↩

Computed for this article in research/code/karmin-filter-matched.ts from the L, SNPs and FS columns of Karmin's Supplementary Table S2.  Tree SNPs per megabase:  filter b 3,887;  c 3,923;  d 3,961;  b+c 3,986;  b+d 3,970;  c+d 3,950;  b+c+d 3,954;  a+b+d 4,045;  a+b+c+d 4,035.  Father-son differences per base, as multiples of a+b+d's 6.92 × 10⁻⁸:  b 14.3;  c 4.37;  d 2.60;  b+c 4.22;  b+d 2.13;  c+d 1.92;  b+c+d 1.90;  a+b+c+d 0.91.  Filter a alone, which keeps 46.4 Mb, shows the same mismatch more strongly:  7.5 times the father-son density and 0.28 times the tree density.  The filters keep overlapping but different stretches of the Y, so each density belongs to the sequence that filter keeps.  On a neutral model the expected number of tree SNPs in a stretch of sequence is its mutation rate times the tree's total length, the same for every stretch, which is why the two densities should move together if the father-son differences were inherited mutations. ↩

Elhaik, Eran et al.  An extended reply to Mendez et al.: The 'extremely ancient' chromosome that still isn't.  2014.  The uncensored version of a reply by four critics of Mendez et al. 2013, written after five of that paper's eleven authors published a rebuttal.  The authors say their earlier reply was censored, and they recompute the age of the A00 split from the raw counts of differences, averaging about 214,500 years.  The article uses it to show that the 338 kya date depends on analytical choices that make it larger.

1. 18a.

   Their recomputation of the A00 divergence from the raw counts:  “one can calculate divergence times of 43/240,000/10-9 → 179,000 years and 45/180000/10-9 → 250,000 years, for an average of 214,500 years, very similar to the TMRCA obtained using a likelihood-based method:  209,500 (95% CI: 168,000–257,400) years.”  The article's table takes the smaller of their two counts, 43 derived differences over 240 kb, and scales it to 10.3 Mb, giving 1,845.  The other count would give 2,575 and a larger required rate, so the conservative choice is used. ↩

2. 18b.

   Quoting private correspondence it reproduces: Thomas Krahn, second author of Mendez et al. 2013 and the person who sequenced the A00 chromosome, wrote:  “Yep, we were extensively discussing this issue.  My numbers were more like [your] 200ky for the A00 split but Fernando [Mendez] insisted using autosomal SNP frequency data for dating, which I thought is absolute nonsense for the Y mutation mechanism.  Anyhow, it was his job to do the dating.”  Elhaik and colleagues are hostile critics of Mendez 2013, and the correspondence is reproduced by them, not by Krahn, so it is reported here as their quotation of him.  Mendez's 338 kya against Krahn's 200 ky is the factor of 1.6. ↩

Hallast, Pille et al.  “The Y-Chromosome Tree Bursts into Leaf: 13,000 High-Confidence SNPs Covering the Majority of Known Clades.”  Molecular Biology and Evolution 32(3):661–673.  Mar 2015.  A sequencing study that built a Y-chromosome family tree from about 13,000 high-confidence SNPs (single-letter DNA differences), covering most of the known branches.  It compares the average number of mutations from the root to the tips in lymphoblastoid cell lines (LCLs), blood and saliva.  The article uses its blood count as a direct count of mutations rather than a model's output.

1. 19a.

   “A comparison of the mean number of mutations to the root of the tree for the three different tissue sources … shows that MSY sequences in the LCLs analyzed here indeed carry significantly more mutations (mean of 471, n = 152; P = 0.00124, one-way analysis of variance) than the sequences from blood (mean of 468, n = 208) or saliva (mean of 466, n = 88).”  The paper's own root TMRCA is "approximately 126 ka" on the Xue pedigree rate, the shallowest modern estimate and still twenty-eight times the Flood.  The 468 blood figure over 3.7 Mb scales to 1,303 over 10.3 Mb, against 1,339 counted directly off Poznik's branch files and 1,490 implied by Poznik's dated tree.  The scaling assumes uniform mutation density, and Hallast's 3.7 Mb excludes the ampliconic and X-transposed regions that Poznik's 10.3 Mb partly includes, so treat 1,303 as an approximation.  The LCL mean of 471 is higher than the blood and saliva means, which is the cell-line artifact Xue 2009 measured directly, so the blood figure is used here. ↩

Computed for this article in research/code/young-clock-checks.ts.  The rate needed is root-to-tip mutations divided by 10.3 Mb and by 4,500 years, then by Helgason's 8.706 × 10⁻¹⁰.  The A00 and CT depths, 1,339 and 565, are counted from Poznik's branch files.  Generation time is granted Swiel's ceiling, a 10% branch shortening.  Patriarchal drive is granted Carter's Model 3, about 55 extra mutations in the first generation on the 10.3 Mb scale, as a share of the 565-mutation CT depth;  the measured paternal-age slope would give 3.86, a factor of 1.007.  Karmin's filter-c rate, 3.02 × 10⁻⁷ per generation, is 10.0 times Helgason's 3.005 × 10⁻⁸. ↩

Carter, Robert W.  “The promises and pitfalls of correlating Y chromosome genetics to human history.”  Journal of Creation 36(2):34–39.  Aug 5, 2022.  A review in a creationist journal of Nathaniel Jeanson's 2022 book Traced: Human DNA's Big Surprise, by a reviewer who says he and Jeanson have discussed these issues in depth.  The review praises the research and calls the book's conclusions perhaps more tentative than its claims, and it describes the book's rule for counting mutations per generation.  The article uses it for Jeanson's per-generation mutation rule and for Carter's warnings about filtering and about Jeanson's choice of root.

1. 21a.

   Summarising Jeanson's method:  “he applies the general rule that three Y chromosome mutations occur each generation, except for the first generation after Noah, to which he applies seven mutations.”  The review has two citation defects.  Its reference 2 gives Jeanson and Holland's Y-clock paper the volume, pages and year of Jeanson's 2020 Amerindian paper, printing "ARJ 13:23–33, 2020", where the correct locator is ARJ 12:393–404, 2019.  Its header dates Traced to 2022, while reference 1 dates it 2021. ↩

2. 21b.

   “It is not trivial to derive mutation rates from sequencing data, because the error rates inherent in the data are on the same order of magnitude of the in vivo mutation rate.  Thus, all papers published on the subject to date have applied significant filtering steps to the data before the final estimates are reported.  This is frustrating, to say the least, since an accurate estimation of the mutation rate is one of the things that will separate the young-earth and the evolutionary models of human history.” ↩

3. 21c.

   “however, and he admits this in the book, this necessitates that some lineages have a higher mutation rate today or at least had a higher mutation rate in the past.  Since the most discordant lineages are in Africa, he appears to be saying that (some) Africans are more mutant than non-Africans.”  Carter's verdict on the book:  “The conclusions are perhaps more tentative than the claims of the book indicate, so readers should be cautious about taking what is said as the final word.”  He also rejects Jeanson's Amerindian result:  “To say that, at random, all lineages but one peter out across both continents is making a statistically indefensible argument.” ↩

Swiel, Yaniv et al.  “Resolving the source of branch length variation in the Y chromosome phylogeny.”  Genome Biology 26:4.  Jan 6, 2025.  A genome study from the Max Planck Institute for Evolutionary Anthropology that compares Y chromosomes from present-day people, ancient modern humans and Neandertals.  It finds that the differences in branch length cannot be explained by demographic or biological processes alone, and that reference bias, the mismatch between a sample and the genome it is aligned to, makes mutations go undetected in highly diverged Y chromosomes.  The article uses it to test and reject the two explanations its opponents need, real differences in mutation rate between lineages and differences in generation time.

1. 22a.

   Section Abstract:  “we analyze Y chromosomes from present-day and ancient modern humans, as well as Neandertals, and show that branch length variation amongst human Y chromosomes cannot solely be explained by differences in demographic or biological processes.  Instead, reference bias results in mutations being missed on Y chromosomes that are highly diverged from the reference used for alignment.” ↩

2. 22b.

   “significantly fewer mutations are identified for those African Y chromosomes that are the most diverged from the reference, and this difference increases linearly with increasing divergence from the reference … We find that ∼52 mutations are missed with each additional pairwise difference from the reference per 10,000 basepairs.” ↩

3. 22c.

   “We find that the branches of the divergent Y chromosome lineages are between 30% and 46% shorter than expected … This suggests that multiple changes in mutation rate are needed to explain the observed branch shortening, and that the mutation rate would need to have been up to 46% slower than that estimated in non-Africans.” ↩

4. 22d.

   “We find that no reasonable change in male generation time could account for the apparent 46% shorter branch for the Neandertal lineage, as a change in male generation time could result in, at most, a 10% branch shortening.” ↩

5. 22e.

   “Although we do not observe any significant haplogroup-specific variation after accounting for the effect of reference bias … a larger Y chromosome dataset with multiple samples representing specific lineages in populations of interest may allow for the detection of subtle branch length differences.” ↩

Carter, Robert W. et al.  “An Overview of the Independent Histories of the Human Y Chromosome and the Human Mitochondrial Chromosome.”  Proceedings of the Eighth International Conference on Creationism, 133–151.  Creation Science Fellowship, Pittsburgh.  2018.  A conference paper from a creationist conference, by two authors at the FMS Foundation and one statistician at the University of Idaho.  It compares Y chromosome and mitochondrial DNA from 1000 Genomes and Simons Genome Diversity Project samples, and concludes that Y Chromosome Noah and Mitochondrial Eve lived less than 10,000 years ago.  The article uses it for the patriarchal drive idea and for the Helgason mutation rate that its authors accept.

1. 23a.

   “Using detailed genealogical knowledge, Helgason et al. (2014) reported a rate of 8.71x10-10 per site, per person, per year for the Y chromosomes of a selection of Icelandic males.  We must point out that modern Icelandic males are hardly an acceptable analogue for all males throughout all history.”  Note the contrast with their treatment of calibrated rates:  “Their figure of 1.7x10-8 mutations per site, per year (or one mutation every 3,624 years) was ridiculously low and is entirely influenced by the assumed human-chimp split time of 6.5 MA.  We reject all phylogenetic mutation rates as unrealistic.” ↩

2. 23b.

   “In the biblical model, the patriarchs grew to be exceedingly old.  The very old patriarchs would have contributed a huge number of new mutations to their children … We note that Noah is the oldest father recorded in the Scriptures.  Because of this, Shem, Ham, and Japheth could have received a huge number of new mutations, and so it is possible that each son could have established, in a single generation, a substantial new branch on the phylogenetic tree.  Also, the paternal age-effect appears to be non-linear (Crow 1997), perhaps even exponential, meaning that very old men having children late in life could have instantaneously created brand new lines on the family tree.” ↩

3. 23c.

   Table 4:  Y haplogroup founder ages, minimum from Jeanson & Carter 2017 and maximum from Xue 2009: A1 6,282 to 106,800 years;  B 4,112 to 69,900;  C 3,526 to 59,950;  D/E 3,331 to 56,625;  G 2,313 to 39,326;  Q/R 2,319 to 39,428;  average 3,110 to 52,900 years.  Their own caption:  “Clearly, one cannot simply apply a linear rate estimate to the distance data, especially since the different branches have had demonstrably different rates of mutation accumulation.”  Section Abstract:  The paper's abstract concludes that the founding individuals.  “lived less than 10,000 years ago, which is most consistent with a biblical timeframe.”  That conclusion rests on the minimum column of this table.  A critic should engage the whole table;  a defender should print both columns. ↩

Kong, Augustine et al.  “Rate of de novo mutations and the importance of father's age to disease risk.”  Nature 488(7412):471–475.  Aug 23, 2012.  A genome study that sequenced the entire genomes of 78 Icelandic parent-offspring trios (a father, a mother and a child) at high coverage, and counted the new mutations each child carried.  It found an average of 1.20 × 10⁻⁸ new mutations per letter of DNA per generation, and about two more mutations for every year of the father's age at conception.  The article uses its slope, which it calls measured, as the starting point for the Y estimate.

1. 24a.

   Section Results:  “The number of mutations increases with father’s age (P = 3.6×10−19) with an estimated effect of 2.01 mutations per year (standard error (SE) = 0.17).”  The regression is on paternal and maternal de novo counts pooled, "as they could not be reliably separated without data from a grandchild," and mother’s age correlates with father’s at r = 0.83, so 2.01 is an upper bound on the paternal-only slope.  The divisor that Mendez et al. do not use is given here.  “Effective genome coverage, computed by discounting regions that have either very low (< ½ genome average) or very high (> 3 times genome average) local coverage, the latter often a symptom of misaligning reads, was estimated to be 2.63 billion basepairs.” ↩

Carter, Robert W.  “Patriarchal drive in the early post-Flood population.”  Journal of Creation 33(1):110–118.  Apr 2019.  A creationist modelling paper that runs a computer simulation of a small post-Flood population, with three different curves for how many mutations a father passes on as he ages.  Under its third curve, a convex guess, the three sons of Noah would each start a branch about 160 mutations long, and the author says the curve cannot be tested at biblical ages.  The article uses it as Carter's own estimate of the mutations the patriarchs could have added, and finds that estimate leaves the tree 24 to 30 times too shallow.

1. 25a.

   In the methods he defines three mutation curves, and says:  “Model 2 is another linear model using the factor of '2 extra mutations per extra year of father's paternity' from Kong et al., but also assumes that only 1% of all mutations will occur on the Y chromosome [y = 1 + (age – 20) x 0.02].  This fixes the problem with Model 1 at young ages, but almost certainly under-represents the mutation rate at high ages.  Model 3 is a polynomial model developed as a best guess of the expected mutation curve [y = 1 + (age – 20)/10 + ((age – 20)/100) 3].”  His Y length is given earlier:  “Given a sequenced portion of the Y chromosome of approximately 30 million bases, this amounts to just under 1 mutation (0.942) per person per generation, in modern Icelandic males.” ↩

2. 25b.

   In the results he writes:  “The initial generation (Noah to Shem, Ham, and Japheth), would have created three massive new branches 160 mutations long (using mutation model 3).  Since the model starts with only three men, 100% of the men carry branches that long at year zero.”  Arithmetic check: 1 + (500 − 20)/10 + ((500 − 20)/100)³ = 159.6, which rounds to his 160.  Model 2 at the same age gives 1 + 480 × 0.02 = 10.6. ↩

3. 25c.

   On the curve's status:  “we have zero opportunity to test it in men with 'biblical' ages of several centuries, meaning we can only speculate about the mutational burden given to us from Noah and his children and grandchildren.”  And in the Discussion:  “patriarchal drive has been shown to be a real possibility and should be incorporated into any basic model of biblical genetics developed in the future … Yet, the results depend strongly on the mutation model, the rate of population growth, and multiple demographic trends we know nearly nothing about.” ↩

Computed in research/code/young-clock-checks.ts.  On the A00 root the compression factor is Karmin's 254,000-year root divided by 4,500, or 56.4, which moves 8,000 to 4,000 years ago to 142 to 71 years ago.  On the CT root it is Poznik's 76,000-year CT split, on an ancient-DNA rate close to Karmin's, divided by 4,500, or 16.9, which moves 8,000 to 4,000 years ago to 474 to 237 years ago, and Zeng's 7,000 to 5,000 to 414 to 296. ↩

Batini, Chiara et al.  “Large-scale recent expansion of European patrilineages shown by population resequencing.”  Nature Communications 6:7152.  May 19, 2015.  A resequencing study of 3.7 Mb of Y-chromosome DNA in 334 European and Middle Eastern men from 17 populations.  Three lineages (I1, R1a and R1b), which make up 64% of the sample, have coalescent times between 3.5 and 7.3 thousand years ago, where coalescent time is when the lineages in the sample share a common ancestor.  Thirteen of the 17 populations share a demographic expansion starting roughly 2.1–4.2 KYA.  The article uses it to show that dramatic recent male founder events happen, with three lineages covering most of the sample.

1. 27a.

   Section Abstract:  “Resequencing of 3.7 Mb of MSY DNA in 334 males, comprising 17 European and Middle Eastern populations, defines a phylogeny containing 5,996 single-nucleotide polymorphisms.  Dating indicates that three major lineages (I1, R1a and R1b), accounting for 64% of our sample, have very recent coalescent times, ranging between 3.5 and 7.3 KYA.  A continuous swathe of 13/17 populations share similar histories featuring a demographic expansion starting ~2.1–4.2 KYA.” ↩

Carter, Robert.  How carbon dating works.  Creation Ministries International.  2021.  An explainer by the young-earth geneticist Robert Carter on how carbon-14 dating works and why creationists hold that the Flood made early samples read too old.  It gives the most careful young-earth estimate of how large that error is at the age of ancient Egypt, a few centuries, and says that no young-earth calibration curve exists.  It matters here because the Saqqaq man's radiocarbon date falls in that range.

1. 28a.

   Section Carbon dating is affected by the strength of Earth’s magnetic field:  “If ancient samples started out with less carbon-14 than we expect, they will appear to be older than they really are. … This age inflation should also increase as you sample older and older objects.  Thus, we might expect samples from ancient Egypt or Jericho to be inflated by a few centuries.  The carbon date of any sample that was formed immediately after the Flood would be even more inflated, perhaps by 50,000 years or more.” ↩

Brown, Robert H.  “Correlation of C-14 Age with Real Time.”  Creation Research Society Quarterly 29(1):45–47.  Jun 1992.  A three-page young-earth paper deriving an equation that converts radiocarbon ages into calendar dates, assuming that after a Flood 5,350 years ago the air's carbon-14 started at about 1% of today's level and rose exponentially.  It prints a conversion table.  It is the only published young-earth radiocarbon conversion with a table, so it shows how far such a model moves a 4,000-year radiocarbon date.

1. 29a.

   Page 46, table Table I:  “In my judgment the most suitable value for F is 5350.”  F is the date of the Flood in years before present.  The table's rows bracketing the Saqqaq date give real-time age T against radiocarbon age R:  T 3700, R 3920;  T 3800, R 4070.  Interpolated, a radiocarbon age of 4,044 is a real age of about 3,780 years. ↩

Computed in research/code/young-clock-checks.ts.  Test 1:  Poznik's pedigree-column split times put CT at 65.0 thousand years and the Q-R split at 30.0, so the split lies 1 − 30.0/65.0 = 0.538 of the way from CT to the present;  the ancient-DNA column gives 0.539.  The mean depth below CT is 565 mutations, so the split is about 304 mutations down, and a constant 4,500-year clock lays down 63 in 500 years.  With the Flood 4,500 years before 2026, in 2474 BC, the split falls 2,423 years later, in 51 BC.  Saqqaq's calibrated range, 4,170 to 3,600 years before 1950, is 2220 to 1650 BC.  On Brown's timescale the split falls 2,469 years before 1950 and Saqqaq at about 3,783, a gap of 1,313 years;  Brown's conversion is applied to the uncorrected radiocarbon age, and a marine-reservoir correction would make Saqqaq a few centuries younger.  Test 2:  19.5 transversions times 2.7, for Karmin's 1.7 transitions per transversion, is 52.65 mutations over 8,819,704 bases.  Over 4,000 years that is 1.49 × 10⁻⁹ per base per year, 1.71 times Helgason's;  over Brown's 3,783 years, 1.58 × 10⁻⁹;  over 3,600 years, 1.66 × 10⁻⁹;  over 3,200 years, a generous allowance for the reservoir correction on Brown's scale, 1.87 × 10⁻⁹.  Jeanson's 3.02 × 10⁻⁷ per generation is 1.01 × 10⁻⁸ per year at 30-year generations, which spreads 52.65 mutations over at most 593 years. ↩

Cruciani, Fulvio et al.  “A Revised Root for the Human Y Chromosomal Phylogenetic Tree: The Origin of Patrilineal Diversity in Africa.”  American Journal of Human Genetics 88(6):814–818.  Jun 10, 2011.  A study that revised the root of the human Y chromosome family tree, the trunk from which all male lineages branch, after adding previously unsampled basal African lineages.  It put the most recent common ancestor of the tree at about 140 KY.  The article uses it for the authors' own statement that a deeper root is easier to reconcile with plausible scenarios of modern human origin.

1. 31a.

   Section Summary:  “An estimate of 142 KY was obtained for the coalescence time of the revised MSY tree, which is earlier than that obtained in previous studies and easier to reconcile with plausible scenarios of modern human origin.”  What drove the deeper root was the discovery of previously unsampled basal African lineages, not a re-fitting of the clock.  The article reads the candour here as quotable, not as circular reasoning. ↩

2. 31b.

   Section Discussion:  “The present figure of about 140 KY for the inferred most recent common ancestor (MRCA) of the MSY phylogeny is older than previous estimates (about 100 KY or below) and easier to reconcile with plausible scenarios of modern human origin.  Clearly, calculation of the precise age of the tree largely depends on the accuracy of the assumed mutation rate.” ↩