Bigfoot DNA: Did It Really “Come Back Human”?

What the publicly available genetic evidence actually shows

The commonly repeated statement that alleged Bigfoot DNA has been tested and shown to be ordinary human DNA is not supported by the public evidence reviewed here.

This is a correction of an overstated claim, not a positive identification. Nothing below is evidence that Bigfoot exists.

What the public record actually contains is:

• some alleged Bigfoot samples that resolve as ordinary animals;
• some samples with failed or unusable DNA recovery;
• Ketchum material whose surviving consensus data are inadequate for the extraordinary claims made from it;
• one publicly deposited alleged-Bigfoot sample — Sykes 25072, GenBank KJ155708 — that yielded human mitochondrial DNA;
• no publicly available, well-provenanced alleged Bigfoot specimen with deep nuclear whole-genome data sufficient to distinguish an ordinary modern human from a genetically unusual human-lineage population.

So "Bigfoot DNA came back human" is too strong. The statement the evidence supports is narrower: an unusual human-lineage explanation remains untested. That is not a hint that such a population exists. It is a statement about a gap in the testing.

The samples

Sykes 25072 — the one real human positive, and its blind spot

The 2014 Sykes et al. study characterised each of its samples with a 104 bp fragment of mitochondrial 12S rRNA (human positions 1093–1196). One sample, no. 25072, returned a human sequence. It is deposited as KJ155708 — Homo sapiens isolate 25072, 104 bp, tissue_type="hair", geo_loc_name="USA: Texas". This is the single publicly verifiable alleged-Bigfoot sample that produced a deposited human sequence.

Three things about it matter, and all three are usually dropped when the result is summarised:

1. 104 base pairs of mitochondrial DNA is a species barcode, not a genome. It identifies a maternal lineage. It carries no information about the nuclear genome, where almost all of an organism's biology — including anything that would make a population phenotypically unusual — actually resides.
2. The follow-up was reported but never deposited. The paper states that hypervariable region 1 (positions 16000–16400) was also sequenced and matched the revised Cambridge Reference Sequence, an ordinary European maternal lineage, which excluded a Neanderthal-type maternal lineage. That sequence is not public. The deposit is 30 records, all 12S: KJ155696–KJ155724 plus KJ607607 — and KJ607607 is Ursus americanus isolate 25202, not the HVR1 of 25072. The HVR1 result exists only as a sentence in the paper.
3. No nuclear sequencing was ever reported for this sample, or for any other sample in the study.

The voucher position is similarly soft. The study states that voucher samples were deposited in the Heuvelmans Archive at the Museum of Zoology, Lausanne. That is an institutional statement about the study as a whole; whether material from this sample survives is unverified, and GBIF holds no record under catalogue number 25072. The museum's own holdings would have to be queried directly, and a deposited voucher does not imply that usable DNA remains in it.

So 25072 establishes a modern-human maternal mitochondrial lineage. It does not establish that the nuclear genome behind that hair was an ordinary modern human genome, because the nuclear genome was never examined. That is the precise blind spot.

Ketchum — the surviving data cannot carry the claim

The 2013 Ketchum et al. paper claimed 111 samples, with nuclear "whole genome" work on three: 26, 31 and 140. No sequence from it was ever accessioned. The archived correspondence shows a single 669 bp mitochondrial submission, BankIt1563470, queried first over its organism name ("unknown hominid") and then over human-subject consent once relabelled Homo sapiens, and finally deleted without an accession. GenBank returns no records for the paper's proposed name.

What survives publicly is three nuclear consensus sequences — 2.73, 0.53 and 2.10 Mb, recoverable from the Wayback Machine — with no reads, no base qualities and no coverage information. They were built, per the paper, in CLC Bio Genomic Workbench 5.1 from a subsample of reads assembled against human chromosome 11 as the reference.

Classifying them independently:

• Sample 26 is consistent with Ursidae at family level (100% identity over 884 bp).
• Sample 140 is consistent with Canidae at family level (100% identity over 448 bp).
• Sample 31 is human-like but uninterpretable. Pooled identity is 98.44%, consistent with single-read consensus error, and the sequence contains essentially no unknown bases across 0.53 Mb — which implies the assembler filled uncovered positions from the human reference rather than from data. A "human" result from a human-reference-guided consensus with no raw reads is not a species determination.

One incidental finding is worth recording because it cuts against a human-lineage reading rather than for it: non-human mammalian sequence still returns Homo as its nearest BLAST hit in conserved regions. A superficial human hit does not establish human origin.

The surviving Ketchum data cannot establish a novel human hybrid or an unusual hominin. They also cannot be used to rule one out. They are simply inadequate.

Walla Walla, 1995 — a failed extraction, not a result

Hair was collected on 5 August 1995 in the Blue Mountains east of Walla Walla, Washington, by Paul Freeman, Bill Laughery and Wes Sumerlin. W. Henner Fahrenbach of the Oregon Regional Primate Research Center described it morphologically as coming from two individuals of one species and as "indistinguishable from human hair by any criterion." DNA work by Paul Fuerst and J. A. Poe at Ohio State University obtained no mitochondrial sequence at all — in their words, their studies "have not yielded a sequenced mitochondrial gene fragment to determine the phylogenetic affiliation of the creature." On 20 March 1998 the co-authors withheld their manuscript pending tissue rather than hair, on the explicit grounds that ambiguous results would be misused by both camps. No sequence exists, so none could be deposited. This is often cited as human-like hair; it is not a human DNA result of any kind.

Zana and Khwit — a calibration case, not a Bigfoot sample

Zana, a woman held in captivity in 19th-century Abkhazia, and her son Khwit are not an alleged Bigfoot specimen. They are included here because they are the only publicly available deep-ish genomes attached to a historical "wild human" account, which makes them a useful test of what this kind of data can and cannot show.

The published findings reproduce. The DNA is authentic ancient DNA, with deamination damage consistent with genuine double-stranded libraries. Zana is female, Khwit male. The mother–son relationship holds. The mitochondrial haplogroup is L2b1b, and the ancestry is modern human.

Beyond that, a bounded anomaly search found nothing compelling. No large copy-number variant was detected at any of the five loci tested for hypertrichosis and gigantism (17q24.2-q24.3, Xq27/SOX3, Xq26.3/GPR101, MSTN, SMAD6). A genome-wide depth sweep produced 15 candidate regions for Zana against Khwit, compared with 13 for a pair of unrelated ancient genomes run through the same code — indistinguishable from noise. Mitochondrially, 16,464 jointly callable positions showed zero discordance between mother and son, and no convincing heteroplasmy beyond ancient-DNA damage patterns.

The honest statement of that result is narrow:

No compelling anomaly was detected in the portions of the deposited Zana/Khwit genomes that could be tested reliably. Several classes of variation remain inadequately testable from these processed low-coverage BAMs.

The limitations are real and specific, not hedging. The deposited files contain no soft-clipped reads and no supplementary alignments, and only 0.06% of reads retain pair information — so split-read and discordant-pair breakpoint detection, and with them inversions and balanced rearrangements, cannot be tested from this deposit at all. Mean depth near 3× rules out reliable genotype and rare-variant calling. And a broader sweep that normalised Zana and Khwit against unrelated ancient controls had to be discarded: the two sets had been aligned against different reference assemblies, one with decoy contigs and one without, which manufactures apparent deletions in exactly the repetitive regions where copy-number variants are usually called.

Zana's genome does not show something strange. It also was not in a position to.

On whether a visually unusual human-lineage population is biologically conceivable

This section exists to bound the question, not to support it. Reviewing human genetic and developmental variation establishes only that:

• extreme body hair can be inherited within Homo sapiens;
• stature, muscularity and craniofacial robusticity vary substantially within modern humans;
• truly ape-like limb proportions would be strongly constraining for a human-lineage population.

That is a statement about the space of possible human phenotypes. None of it is evidence that Bigfoot exists, and none of it is evidence about any specific sample. It matters only because it means a "human" barcode result cannot by itself close the question — which is why the missing nuclear data is a gap rather than a formality.

Conclusion

The public genetic record does not show that Bigfoot is human. It also does not show that Bigfoot is an unknown ape, a hybrid, or an archaic hominin.

What it shows is that the often-repeated claim that "Bigfoot DNA was tested and came back human" compresses weak, incomplete and heterogeneous evidence into a conclusion the underlying data do not support. One deposited human mitochondrial barcode of 104 base pairs, an undeposited follow-up, two family-level animal identifications, one uninterpretable reference-guided consensus, and a failed extraction do not add up to a tested nuclear genome.

The decisive experiment would require a well-provenanced biological specimen, preserved raw material, deep nuclear sequencing, publicly released raw reads, and appropriate human and non-human controls. No specimen meeting that standard was identified in this investigation.

Methodology, discarded analyses, the Zana/Khwit sweep in full, and a 154-row log of approaches that were tried and rejected are held in the investigation's internal research notes rather than here.