Plant mapping populations are usually created from F1 lines that are derived from two parents that show differing phenotypes for a target trait; for example, a broccoli line that displays rapid post harvest yellowing and a line that stays green longer than other broccoli lines.

Parent 1 x Parent 2 ----- F₁ progeny

A segregating F2 population is formed from the F1 individuals either by randomly crossing these individuals or allowing them to self.

F₁ -------- F₂

The F₁ individuals will be heterozygous these lines are also used to produce doubled haploid lines (DH lines) using microspore culture. Haploid plantlets produced from embryos usually undergo spontaneous doubling producing DH lines (this can be achieved using colchicine; colchicine binds to tubulin and prevents microtuble polymerisation, acting as a mitotic poison, pretty potent for a plant derived [autumn crocus] secondary metabolite). Doubled haploid lines are advantageous since they are homozygous or fixed lines, and as such can be replicated in trials allowing better estimates of within and between line variation. Plus, they can be produced in a relatively short period (1-2 yrs).

Homozygous (fixed) lines can be produced from F₂ individuals using single seed descent. One seed is harvested from each F₂ individual - grown into an F₃ plant - a single seed is harvested and grown into an F₄ plant and so on until F₈ to F₁₀ generations were the lines will be virtually homozygous at all loci. Pockets of heterozygocity sometimes persist, however if these are identified, more rounds of selfing may eliminate these. The end result is a set of recombinant inbred lines (RILS). This can take up to 5 years depending on the species.

If lines are intercrossed (sibling mating) in the F₂, F₃ and F₄ generations Advanced Intercross Lines (AILs) can be generated, these are then maintained by single seed descent.

Fixed lines can be replicated in trials allowing for better estimates of trait means, with reduced environmental variation; this is important for estimates of trait heritability and QTL analyses. Having fixed lines is essential for genotype evaluation.

Members of a fixed mapping population will contain differing amounts of recombination and linkage disequilibrium between loci will be present. The population is genotyped with molecular markers and linkage analysis performed to estimate a linkage map for the population.

Once a particular line or lines have been selected based on the presence of a favorable genotype linked to a trait of interest, e.g. a broccoli line that stays green longer than the good parent, this line (donor) can be backcrossed to the poor (recurrent) parent to produce a backcross (BC₁) generation. Subsequent rounds of backcrossing to the poor parent in combination with the use of genetic markers that allow selection of only lines containing the favorable genotype (foreground selection), in combination with markers to select for the poor parent genotype elsewhere in the genome (background selection) allow lines to be created that contain a small region of the beneficial genotype in virtually homogenous background. The resulting progeny are called Advanced Backcross lines (AB lines) or Backcross Inbred Lines (BILs).

Depending upon the donor parent, this method can be used to generate a population that have introgressed segments of the good parent in the background of the poor parent, that overlap to provide coverage across all chromosomes. If a RIL was used than the resulting progeny may be classed as Near Isogenic Lines (NILs); in this case the desired population will consist of lines that contain small overlapping segments of a particular target region, for example a QTL interval.

Another population type, the natural diversity set, aims to capture as much of the heterogeneous wild species gene pool as possible. In these populations, historical or residual linkage disequilibrium persists between the representative accessions due to the common ancestry/ evolutionary history. In combination with SNP markers these resources have the potential to offer improved resolution for association analysis compared to other mapping populations owing to the historical recombination events captured, for example, DH populations only go through one round of recombination; RILs on the other hand go through many rounds of recombination before becoming fixed.

TILLING (T_arget I_nduced L_ocal L_esions In G_enomes) populations are also common as an exploratory tool for trait improvement; however, these are not strictly a mapping population. TILLING populations are mutagenised populations that allow for the identification of point mutations within genes of interest following a screening procedure. The VeGIN lettuce TILLING population is a good example of this type of resource.