Multi-objective optimization in systematic conservation planning and the representation of genetic variability among populations

Biodiversity crises have led scientists to develop strategies for achieving conservation goals. The underlying principle of these strategies lies in systematic conservation planning (SCP), in which there are at least 2 conflicting objectives, making it a good candidate for multi-objective optimization. Although SCP is typically applied at the species level (or hierarchically higher), it can be used at lower hierarchical levels, such as using alleles as basic units for analysis, for conservation genetics. Here, we propose a method of SCP using a multi-objective approach. We used non-dominated sorting genetic algorithm II in order to identify the smallest set of local populations of Dipteryx alata (baru) (a Brazilian Cerrado species) for conservation, representing the known genetic diversity and using allele frequency information associated with heterozygosity and Hardy-Weinberg equilibrium. We worked in 3 variations for the problem. First, we reproduced a previous experiment, but using a multi-objective approach. We found that the smallest set of populations needed to represent all alleles under study was 7, corroborating the results of the previous study, but with more distinct solutions. In the 2nd and 3rd variations, we performed simultaneous optimization of 4 and 5 objectives, respectively. We found similar but refined results for 7 populations, and a larger portfolio considering intra-specific diversity and persistence with populations ranging from 8-22. This is the first study to apply multi-objective algorithms to an SCP problem using alleles at the population level as basic units for analysis.

Development and characterization of new microsatellites for Eugenia dysenterica DC (Myrtaceae)

Microsatellite markers were developed for population genetic analyses of the Neotropical tree Eugenia dysenterica DC (Myrtaceae), after construction of a shotgun genomic library for microsatellite discovery. Nine primers were designed, of which 5 yielded amplified product. These primers were polymorphic for 97 individuals collected in 3 distinct localities. The number of alleles per locus (primer) ranged from 3 to 11 and expected heterozygosities varied from 0.309 to 0.884. The probability of locus identity was ~1.88 x 10(-4) and the probability of paternity exclusion was ~0.9367. The 5 microsatellite primer pairs may be suitable for population genetic studies such as parentage and fine-scale genetic analyses of this species.

Transferability and characterization of microssatellite loci in Anacardium humile A. St. Hil. (Anacardiaceae)

Microsatellite markers were transferred from the cashew, Anarcadium occidentale, to Anacardium humile (Anacardiaceae), a Neotropical shrub from the Brazilian savanna, that produces an edible nut and pseudo-fruit. We tested 14 microsatellite primers from A. occidentale on A. humile. Polymorphism of each microsatellite locus was analyzed based on 58 individuals from three populations. Twelve loci amplified successfully and presented 2 to 9 alleles; expected heterozygosity ranged from 0.056 to 0.869. These 12 microsatellite loci provide a new tool for the generation of fundamental population genetic data for devising conservation strategies for A. humile.

Genetic characterization of natural populations of the medicinal plant Palicourea coriacea (Rubiaceae) with molecular markers

Palicourea coriacea (Rubiaceae) is a herbaceous, perennial species typical of the Cerrado; it is popularly known as "douradinha", because of its yellow flowers. It is utilized in popular medicine, mainly for the treatment of renal diseases. We used RAPD markers to evaluate the genetic structure of nine natural populations of P. coriacea, totaling 168 individuals, collected in the States of Goiás and Bahia. This species showed a high level of genetic diversity, with He values varying between 0.259 and 0.338, with an overall mean of 0.296. Analysis by AMOVA revealed that 23% of the total variability was between populations and 77% was within populations. The estimate of apparent gene flow (Nm) was 0.83. Analyses of the fixation index (f ) using a Bayesian approach yielded a mean value of 0.98, suggesting that this is an autogamous species. Analyses of genetic divergence and spatial pattern of the populations utilizing theta(B) and Phi(ST) matrices, pair to pair, revealed no correlation between geographic distance and genetic distance; the nine populations grouped randomly, without relation to their geographic origin. The hypothesis that geographically close populations should be genetically close was discarded based on the Mantel test; the correlation was 0.155 (P = 0.23). The degree of interpopulational differentiation was relatively high, which allows us to recommend a strategy of sampling for the ex situ conservation of genetic variability, utilizing a larger number of populations. For in situ conservation, we suggest preservation of a larger number of areas in the Cerrado, where this species naturally occurs.

Transferability of short tandem repeat markers for two wild Canid species inhabiting the Brazilian Cerrado

The maned wolf (Chrysocyon brachyurus) and the crab-eating fox (Cerdocyon thous) are two wild-canid species found in the Brazilian Cerrado. We tested cross-amplification and transferability of 29 short tandem repeat primers originally developed for cattle and domestic dogs and cats on 38 individuals of each of these two species, collected in the Emas National Park, which is the largest national park in the Cerrado region. Six of these primers were successfully transferred (CSSM-038, PEZ-05, PEZ-12, LOCO-13, LOCO-15, and PEZ-20); five of which were found to be polymorphic. Genetic parameter values (number of alleles per locus, observed and expected heterozygosities, and fixation indices) were within the expected range reported for canid populations worldwide.