Preparing the "datafile" for LINKAGE: Page 1

Notes from the authors

  • Besides the "datafile", which contains descriptions of loci, you also need to prepare a "pedfile" with pedigree and genotypic information. This interface does not cover the preparation of a "pedfile".
  • This interface is currently in beta testing. Please report any bugs.
  • If you plan to use this interface for your research in the future, please acknowledge the reference: W. Li, V.F.G. Haghighi, T.C. Matise, J. Ott, "Using web browser as the graphical user interface for linkage analysis", American Journal of Human Genetics, supplement, 59, #1782 (1996)
  • This page is copyrighted by Laboratory of Statistical Genetics of Rockefeller University. Request for mirror, copy, reproduction, or collaboration on an improved version should be sent to us.

A summary of all questions to be asked


Q1. Which LINKAGE program do you want to run:
Note:
MLINK calculates LOD score table (it can also calculate the risk), usually for two-point (1 marker plus 1 trait);
LINKMAP/CMAP performs multipoint linkage analysis for general/CEPH pedigrees with fixed map distances;
ILINK/CILINK estimates recombination fractions and max LOD score for general/CEPH pedigrees;
LODSCORE/CLODSCORE is almost identical with ILINK/CILINK, but for 2 groups of loci cases only.

Q2. How many loci?

Note: The number of loci could include one disease locus (if there is one).

Q3. Do you want to calculate genetic risk? yes, no

Note: you can only calculate risk for the trait locus, and the person whose risk is to be calculated should have an unknown trait (e.g. unknown affection status). You will have to identify this person when you prepare "pedfile" using MAKEPED .

Q4. Are you going to use markers on X chromosome? yes, no

Note: another way to ask this question is whether the linkage analysis is "sex-linked".

Q5. Allow new mutation? yes, no

Note: it is equivalent to ask: whether you allow the possibility that a disease allele is not transmitted from the parents, but appearing by a mutation?

Q6. Do you assume "linkage equilibrium"? yes, no

Note: linkage equilibrium between two alleles at two loci means the absence of statistical correlation between the two alleles. If it is the case, the frequency for a particular haplotype is simply the product of two allele frequencies.

Q7. Do you assume the recombination fractions in male and female are same, differ by a constant ratio differ arbitrarily?

Note: is real life, the recombination fractions in male and female are usually different. But assuming the two to be equal simplifies the calculation.

Q8. Do you assume interference between two crossover events? no
(Only when the number of loci in Q2 is 3, can you choose the "yes" answer below!!)
yes, but no specific form, yes, with a specific form