'Statistical Sudoku' Challenge: splitting the Penrose frequency table
 
This table is from the 1933 article 'The relative effects of paternal and maternal age' in Down's syndrome by the geneticist Lionel Penrose.
 
The row totals at the right of the table contain the paternal-age-specific frequencies for the 'Normal' (N) and Down's children (D) separately.
Likewise, the columns totals at the bottom contain separate maternal-age-specific frequencies for the 'Normal' and Down's children.
 
However, the frequencies in the body of the table are the sums of the frequencies of Normal and Down's cases.
 
The CHALLENGE is to split each row-column-specific sum into the frequency of N and the frequency of D cases.
Table in Penrose, 1933
It is easy to split a few of the small (N + D) counts into their separate N and D components. e.g. N[18,18] = 1; D[18,18] = 0.
But for larger (N + D) counts closer to the centre of the table, the task of splitting them into N and D becomes progressively more difficult.
 

Background
 
Down syndrome or Down's syndrome, also known as trisomy 21, is now known to be a genetic disorder caused by the presence of all or part of a third copy of chromosome 21. It is typically associated with physical growth delays, characteristic facial features, and mild to moderate intellectual disability.
 
The article The relative effects of paternal and maternal age in Down's syndrome by Lionel Penrose, was published in the Journal of Genetics, May 1933, Volume 27, Issue 2, pp 219-224. This was before its genetic nature was known. Penrose was one of the first to establish that its frequency rate was a stronger function of maternal than paternal age, and (in a separate article in 1934) of maternal age than birth order  

JH became aware of Penrose's work through the Wellcome Library's recent digitization of The Lionel Penrose papers. These papers also contain extensive correspondence with statistician Ronald Fisher, a pioneer in statistical genetics.
 
Lionel Penrose was the father of mathematician Roger Penrose, chess Grandmaster Jonathan Penrose, mathematician Oliver Penrose (who graciously provided JH with this reconsideration of his father's 1933 work. [It appears as the chapter 'A beautiful method of analysis': O. Penrose, pp 443-451 in "Fifty years of human genetics: A Festschrift and liber amicorum to celebrate the life and work of George Robert Fraser", edited by Oliver Mayo and Carolyn Leach, ISBN 9781862547537, Wakefield Press, Adelaide, 2007]), and geneticist Shirley Victoria Hodgson (who graciously sent JH a copy of her father's book "Clinical and Genetic Study of 1280 Cases of Mental Defect: Colchester Survey").
 


James Hanley,
Dept of Epidemiology, Biostatistics and Occupational Health,
McGill University
 
jhanley.biostat.mcgill.ca




Updated: February 2, 2015