To the Editor: In a recent article1 in the American Journal of Human Genetics, the authors reported the use of scan statistics for identifying a cluster of variants associated with autism. We applaud the authors for their careful work but would like to point out that scan statistics as a mapping tool are not new in human genetics.
Although scan statistics have been known since the 1960s (Joseph Naus, “father of the scan statistic”2), the first idea for their use in human genetics was based on the observation that true peaks in linkage scans tend to be wider than false peaks.3 It was this idea that prompted us to introduce scan statistics into human gene mapping.4 For this scenario, it was shown that the power gain of scan statistics over usual LOD-score analysis is small.5 However, it has been noted that in the presence of multiple susceptibility variants within a short region, scan statistics can greatly improve power.6 It was this latter situation that the authors addressed in their article1 and that had previously prompted us to implement the scan-statistics approach in a freely available computer program, scanstat, which has been on our website for many years. Twenty years ago, scan statistics were also proposed for sequence analysis.7
Web Resources
The URL for data presented herein is as follows:
Scan Statistics in Gene Mapping, http://lab.rockefeller.edu/ott/programs
References
- 1.Ionita-Laza I., Makarov V., Buxbaum J.D., ARRA Autism Sequencing Consortium Scan-statistic approach identifies clusters of rare disease variants in LRP2, a gene linked and associated with autism spectrum disorders, in three datasets. Am. J. Hum. Genet. 2012;90:1002–1013. doi: 10.1016/j.ajhg.2012.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wallenstein S. Joseph Naus: Father of the Scan Statistic. In: Glaz J., Pozdnyakov V., Wallenstein S., editors. Scan Statistics: Methods and Applications. Birkhäuser Boston; Boston: 2009. pp. 1–25. [Google Scholar]
- 3.Terwilliger J.D., Shannon W.D., Lathrop G.M., Nolan J.P., Goldin L.R., Chase G.A., Weeks D.E. True and false positive peaks in genomewide scans: Applications of length-biased sampling to linkage mapping. Am. J. Hum. Genet. 1997;61:430–438. doi: 10.1086/514855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hoh J., Ott J. Scan statistics to scan markers for susceptibility genes. Proc. Natl. Acad. Sci. USA. 2000;97:9615–9617. doi: 10.1073/pnas.170179197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Siegmund D. Is peak height sufficient? Genet. Epidemiol. 2001;20:403–408. doi: 10.1002/gepi.1009. [DOI] [PubMed] [Google Scholar]
- 6.Hernández S., Siegmund D.O., de Gunst M. On the power for linkage detection using a test based on scan statistics. Biostatistics. 2005;6:259–269. doi: 10.1093/biostatistics/kxi007. [DOI] [PubMed] [Google Scholar]
- 7.Karlin S., Brendel V. Chance and statistical significance in protein and DNA sequence analysis. Science. 1992;257:39–49. doi: 10.1126/science.1621093. [DOI] [PubMed] [Google Scholar]