I have already googled it before but yet unsuccessful in finding the whole topic with intuition. Please help.
A guide to an associated book would be greatly valued.
First you need two, hopefully homozygous, strains that differ in some measurable quantitative trait. In worms it has been done with lifespan; in mice injected with ethanol it has been done with long-sleep and short-sleep periods. The two strains need to differ in their RFLPs. For flies there are probably lots of behavioural traits one could use.
You set up a cross, and then mate individual brothers to sisters, and keep repeating this breeding pattern for multiple generations to establish recombinant inbred lines. After 10 generations they are basically homozygous, and you take the individual lines and assay them first for the trait you are scoring, and second for the molecular markers. The results of the molecular mapping let your draw the haplotype along each chromosome, so you know which parental strain contributed which portion or each chromosome.
The results of each quantitative trait scoring let you classify each recombinant inbred line into one of the two parental phenotype classes (or if it is a multi-locus trait then you may get some intermediate lines).
With those two results you will be able to correlate the absence, or presence, of molecular markers along the chromosomes, with the quantitative trait you are studying.
If it works you will be able to narrow in on a genomic region that co-segregates with the trait. If there are multiple genes involved you may be able to assign a probability or LOD score to different regions.