To reconstruct the detailed demographic history of each red panda population, we applied the simulation PSMC v0.6.4-r49 (41) to the whole diploid genome sequences, with the following set of parameters: -N 30 –t 15 –r 5 -p 4 + 25*2 + 4 + 6. We excluded sex-chromosome sequences of the red panda genome by aligning the red panda genome with the dog genome. We selected two to three high-depth sequenced individuals from each population for PSMC analysis (table S3). We estimated the nucleotide mutation rate of red panda using ferret as the comparison species and the following formula: μ = D × g/2T, where D is the observed frequency of pairwise differences between two species, T is the estimated divergence time, and g is the estimated generation time for the two species (42). In this study, the generation time (g) was set to 6 years (26), the estimated divergence time was set to 39.9 Ma ago (15), and D was estimated to be 0.10558. On the basis of the above formula and the corresponding values, a mutation rate of 7.9 × 10−9 mutations per site per generation was estimated for the red panda. In addition, we performed BSP analyses based on mitochondrial genomes of 15,994 bp for two species separately, using BEAST v1.8.2. The best substitution model of HKY + I was selected by ModelGenerator v0.85. A strict clock rate was selected with a nucleotide substitution rate (43) of 1.9 × 10−8. A total of 8 × 108 iterations were implemented with 10% burn-ins. The BEAST running results were assessed, and the BSP plots were produced by Tracer v1.5.

We used the flexible and robust simulation-based composite-likelihood approach implemented in Fastsimcoal2 v2.5.2.21 (44) to infer species/population divergence and demographic history with the following parameters: -n 100000 -N 100000 -d -M 0.001 -l 10 -L 40 -q --multiSFS -C10 -c8. Because of the memory limit of Fastsimcoal2 running, we selected 55 individuals among 65 red pandas for simulation analysis (table S2). Four alternative population divergence and demographic models were explored. For each model, we ran the program 50 times with varying starting points to ensure convergence and retained the fitting with the highest likelihood. The best model was selected through the maximum value of the likelihoods. Parametric bootstrap estimates were obtained on the basis of 100 simulated data sets (table S9). In addition, we performed population-level admixture analysis for detecting gene flow among genetic populations using the TreeMix method (45) with the following running parameters: treemix –bootstrap –k 1000 –se –noss –m 1~5.

Note: The content above has been extracted from a research article, so it may not display correctly.

Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.

We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.