GAW15 Data

GAW15 will include the following three problems:

• PROBLEM 1—Microarray expression profiles (Vivian Cheung & Richard Spielman):
  
  Data are provided for fourteen 3-generation CEPH Utah families (~8 offspring per sibship, ~14 individuals per family). Phenotypes include expression level of genes in lymphoblastoid cells of these family members, obtained using the Affymetrix Human Focus Arrays that contain probes for 8,500 transcripts. For 3,554 of the 8,500 genes tested, Morley et al. (Nature 430:743-747, 2004) found greater variation among individuals than between replicate determinations on the same individual. These 3,554 expression phenotypes (expressed genes) were chosen for further analyses (linkage, etc.) and have been provided to GAW15. For ~100 individuals, array hybridizations were performed in duplicate. The Affymetrix CEL files for all array hybridizations are provided. These files can be read using the Affymetrix MAS software or other software such as dCHIP (http://biosun1.harvard.edu/complab/dchip/) or BioConductor (http://www.bioconductor.org/). Genotypes for members of the 14 families are provided for 2,882 autosomal and X-linked SNPs. The genotypes were generated by The SNP Consortium (http://snp.cshl.org/linkage_maps/).
  
  This data set provides the opportunity to develop and apply methods for simultaneous analysis of a variety of related traits. Natural variation in gene expression is a new idea, and this collection is unique in providing such a large number of phenotypes in a family study.
  
  
  
  
  
  
  

• Problem 2—Rheumatoid arthritis (Chris Amos and collaborators):
  
  Data from the North American Rheumatoid Arthritis Consortium study (NARAC) and from collaborators in Canada, France, and England. The goal of these studies is to understand the etiology of rheumatoid arthritis. It is highly likely that multiple interacting loci influence risk for disease, as evidenced by the considerably higher recurrence risk to siblings as opposed to more distant relatives for this disease. The availability of some known causal loci could facilitate gene-gene interaction studies in these data sets.
  
  

  NARAC:	Available data from NARAC include microsatellite scans and SNP scans for 512 multiplex families. About 90% of the families are Caucasian. Illumina performed analysis of 5744 genome-wide SNPs on all NARAC families as well as 60 families from Canada. NARAC candidate gene data: Data from the PTPN22 locus are available for the NARAC data sets and have been provided by Celera DX from a study of 1519 controls and 1393 cases. The controls are unrelated individuals from a New York City population. Data on candidate genes are also provided by Robert Plenge for 855 unrelated controls and 839 cases. The cases include affected sibling pairs. Candidate genes that have been studied include PTPN22, CTLA4, TNFRS1 and PADI4, among others.   NARAC association mapping: A dense panel of 2300 SNPs was genotyped by Illumina for an approximately 10 kb region of chromosome 18q that showed evidence for linkage in the U.S. and French linkage scans. These markers were individually genotyped on 460 cases and 460 controls. Controls were recruited from a New York City population.   NARAC quantitative phenotypes and clinical measures: Data are provided for two quantitative phenotypes that are used for identifying RA affected individuals: anti-CCP and Rheumatoid Factor IgM. Clinical measures that are available include age at clinical presentation, sex, ethnicity, and presence of erosions. Smoking increases risk for rheumatoid arthritis, and limited smoking data are available for families and controls. There are also individual items reported by physicians of elements that were considered in making a diagnosis of rheumatoid arthritis, which include presence of morning stiffness, swelling of three or more joint groups, swelling of hand joints, symmetrical swelling, and presence of subcutaneous nodules, as well as rheumatoid factor positivity and radiological changes.
  Canada:	The Canadian group provided 60 families that have been genotyped using the Illumina platform used by NARAC (performed at the same time as the NARAC study) as well as 79 families that were genotyped using an Affymetrix 100K platform.
  ECRAF (France):	The European Consortium on Rheumatoid Arthritis Families has provided high-density microsatellite data from 88 families including 105 sib pairs typed with 1,089 microsatellite markers.
  UK:	The UK group provided data from analysis of 10,156 SNP markers genotyped on 157 families. Microsatellites were used in an initial screen of the entire genome using 369 markers that were analyzed on 175 families. A second screen was performed on 197 families using 89 markers in regions that showed evidence for linkage in the first screen.

  
  
  
  
  
  

• PROBLEM 3: Simulated data (Michael Miller):
  
  100 replicates of simulated data are provided, modeled after the rheumatoid arthritis data. Each replicate includes 1500 nuclear families of size 4 (2 parents and an affected sib pair (ASP)) and 2000 unrelated controls. Three sets of autosomal markers are included: (1) a set of 730 microsatellite markers spaced on average 5 cM apart; (2) a set of 9,187 SNPs distributed on the genome to mimic a 10K SNP chip set; and (3) a very dense map of 17,820 SNPs on chromosome 6 (an average inter-marker spacing of 9,586 bp). The data include map information, with lists of markers and their locations, and simulated family, marker, and phenotype data. Phenotype/covariate data include rheumatoid arthritis affection status, age at ascertainment, lifetime smoking, anti-CCP, IgM, severity, age at onset, and age at death. HLA DR genotype also is provided.
  
  

  “Answers” will be provided on request. If you request the answers, we will record the date that you requested them. We ask that you not give the answers to, or get them from, anyone else.