Computational protocol: A Trans-Amazonian Screening of mtDNA Reveals Deep Intraspecific Divergence in Forest Birds and Suggests a Vast Underestimation of Species Diversity

Similar protocols

Protocol publication

[…] Birds were captured in the field using mist-nets at various localities in Ecuador (between 1999 and 2004) and French Guiana (between 2007 and 2008) (, ). Each individual captured was identified, photographed, and a blood sample was collected by venipuncture for genetic analysis. Species selected for the study were those whose Amazonian distribution included both Ecuador and French Guiana, and we included trans-Andean samples from western Ecuador when available. Forty species were sampled in both east Ecuador and French Guiana, and twelve of them were also sampled in the west-Ecuadorian Chocó. Three additional species with more restricted ranges were included in the analysis for comparative purposes: two sister species within the flycatcher genus Mionectes (M. oleagineus and M. macconnelli, the former widely distributed in Amazonia and the latter restricted to the Guianan Shield), and Thalurania fannyi, a hummingbird species restricted to the Chocó but closely related to T. furcata, one of the species used in the trans-Amazonian comparison. In total we obtained sequences from 758 individuals (, ). We follow Restall et al. for species taxonomy and nomenclature.DNA was extracted from blood by membrane purification with 96-well glass fiber filtration plates (Acroprep 96 Filter Plate, 1.0 µm Glass, PALL Corp.) . Cycle conditions were based on previous analyses . A stable segment of ∼910 bp of the cytochrome oxidase I gene was amplified with primers LTyr (TGTAAAAAGGWCTACAGCCTAACGC, Oliver Haddrath, pers. com.), and COI907aH2 (GTRGCNGAYGTRAARTATGCTCG, ). Amplified products were purified by excising bands from the agarose gel and filtering each through a filter tip , then sequenced in both directions using an automated sequencer ABI 3730 (Applied Biosystems), according to the manufacturer’s suggested protocols. The primers used in the sequencing reaction were Ltyr and COI748Ht (TGGGARATAATTCCRAAGCCTGG, ), except for Platyrinchus coronatus, for which COI907aH2 was used as the reverse primer instead.We aligned sequences using Sequencher 4.1.4 (Gene Codes) and Geneious 5.3.6 (Biomatters), and polymorphisms were checked visually for accuracy. Sequences have been deposited in GenBank, and original trace files are available in the BOLD project “Neotropical Birds” ( We calculated sequence divergence using a Kimura-two-parameter (K2P) model of sequence evolution and corrected distances across the Amazon (trans-Amazonian) and across the Andes (trans-Andean) for intra-population polymorphism using Arlequin 3.1 . All samples within each of the three main regions (French Guiana, East Ecuador and West Ecuador) were grouped together to estimate intra-population polymorphism. To examine phylogenetic relationships among haplotypes we constructed haplotype networks for each species using the median-joining algorithm in the program Network 4.6.0 (Fluxus Technologies Inc.). Statistical significance of network branches was estimated by 1000 bootstrap replicates on neighbor-joining trees generated for each species in MEGA 5.0 .We calculated genetic diversity indices and demographic history parameters for Amazonian species with sufficient sample sizes (n ≥6 in both Ecuador and French Guiana). We calculated haplotype and nucleotide diversity indices in DnaSP and tested for rapid changes in population size indicating past population expansions using Fu’s test of neutrality and calculated values of Fs in Arlequin 3.1 .We estimated phenotypic differences among intraspecific lineages by scoring the degree of plumage color divergence in at least one of the sexes, using a four-code key and the following arbitrary criteria: (1) no apparent differences in plumage and no subspecific designations in current taxonomy; (2) slight but diagnosable differences in color shade, intensity or extension, but no difference in the color itself or in color patterning; (3) marked differences in color shade, intensity or extension, but no differences in the color itself or in color patterning; (4) marked divergence in color or patterning. To standardize scoring as much as possible, we restricted the sources of subspecific phenotypic information to the color plates and verbal descriptions in Restall et al. and our own photographic vouchers. Summary descriptions of phenotypic differences on which scores were based are provided in in the Supplementary Materials. To test for an association between genetic distances and phenotypic scores we used a model II simple linear regression with a major axis (MA) regression method as implemented in the lmodel2 package in R 2.10.1 (R Development Core Team).The plumage divergence scores used here represent a very coarse estimate of differentiation, and were produced with two main objectives in mind: (1) showing overall patterns of genetic-phenotypic association (or lack thereof), and (2) identifying potential candidate lineages for new species designation, with the understanding that detailed phenotypic (including plumage, song, and behavioral data) and molecular analyses will be necessary before final taxonomic decisions are made. We realize that for taxonomic purposes, proper analysis of variation in plumage color should be based on more quantitative measures of color (such as those provided by spectrophotometry) taken on study skins rather than qualitative assessments extracted from color plates, verbal descriptions and subspecific designations in Restall et al. , although the latter are of course the result of extensive examination of study skins. We are also aware of the fact that human perception likely underestimates color differences perceived by the avian eye , , yet because most avian alpha taxonomy has been based on qualitative, subjective assessments of differences in plumage color as perceived by human taxonomists, we feel the method is justified within the confines of the stated objectives. […]

Pipeline specifications

Software tools Sequencher, Geneious, Arlequin, MEGA, DnaSP
Applications Phylogenetics, Population genetic analysis
Diseases Pulmonary Fibrosis