Census: do papers state the cell seeding density at which colonies were counted?

Answer

The census yields several counts because the verified submissions use varying definitions of "stated". Under the loosest definition (any explicit seeding or plating cell number for at least one compared arm), 8 of 9 seed-list human papers state a density b9cee624; adding Drozd 2015 gives 9 of 10 f0efd4eb, and adding Steichen 2014, a negative, gives 9 of 11 92911200 (read 2026-09-20). Under a strict definition (density printed for the culture in which colonies were counted, all compared arms determinable) the count is 3 of 9 f13a7146; a number printed for every compared arm gives 4 of 8 human papers fc686757. The 2026-09-07 scan figure of 2 of 10 is not reproduced under any definition here, though the strict human-only count of 2 of 8 is close to it f13a7146.

Summary

The tallies are not matched measurements: they count different things (any number, every arm, colony-count culture) over different denominators (8, 9, 10, 11). I therefore do not treat them as conflicting; the corpus holds no matched pair. Only Manzini 2015 is a clear "not stated" at the colony-forming step among the seed list b9cee624. Steichen 2014 gives only "approximately 20 colonies per 10^4 parental fibroblasts" with no vessel or area, so it is negative in the census units 92911200. Where density is stated, arms are often at unequal densities: Okumura 2019 varies it by design (500 to 4000 cells/well) and Schlaeger 2015 tests 50 k, 100 k and 200 k b9cee624. Equal density across arms is explicit in Lam 2022 (3 x 10^5 per well), Simara 2014 Table 1 and Drozd 2015 (8x10^4 per well of a six-well plate) b9cee624``f0efd4eb. Units (cells/well, per 10-cm dish, per transfection) are not interchangeable.

Evidence

• Warren 2010 (PMC3656821) — Table 1 "Cells plated": 300,000 (BJ), 50,000 (4F vs 5F), 100,000 (RNA vs virus); same starting number in both RNA and virus arms, with well fractions of 1/6 and 1/3. 6cf62a35 f13a7146
• Xu 2013 (PMC3617278, mouse MEFs, outside the human denominators) — "MEFs were reseeded at a density of 5000 cells/well onto 96-well plates"; a dedicated experiment used 5000, 1000 and 200 cells/well, and which factor set gave more colonies changed with density. 613e7763
• Drozd 2015 (PMC4515927, added via paper search) — default 8x10^4 per well of a six-well plate across donor cell types, "unless stated otherwise". f0efd4eb
• Steichen 2014 (PMC4039453, added) — per-10^4-fibroblast basis only; not stated in required units. 92911200
• Seed-list rows (Schlaeger, Trevisan, Okur, Lam, Okumura, Kuebler, Manzini, Simara) — quotes, units and sections searched are in b9cee624; Trevisan and Kuebler are partial. Simara 2014 was read via an open publisher PDF there, while two other submissions did not examine it f13a7146``fc686757.

Methodology

I did not pool the tallies. I compared their stated inclusion rules: any number for at least one arm (8 of 9; 9 of 10; 9 of 11), a number for every arm (4 of 8), and a strict colony-count-culture criterion (3 of 9). Xu 2013 is treated as outside the human denominator, as most submissions did. Rows come from the verified submissions; I did not re-read the papers myself.

Limitations

• Denominators and paper classifications vary between submissions (for example Warren 2010 and Manzini 2015), so the corpus does not settle one count.
• Some rows rest on retrieved passages, so "not found" does not prove absence fc686757``f13a7146.
• The sample is a seed list plus two additions, not a random sample of the literature; it supports "of the N papers examined, M report it" and nothing wider.