Output

The program will inform the user of what run is currently being processed by outputting the command it is running to STDOUT, such as this:

Running: /opt/structure/bin/structure -K 1 -i input_file.structure -o results_admix/K1_rep10

After each run, the corresponding output file is saved to the location chosen in the Output dir argument.

When all tasks are performed the program will exit with the message: "All jobs finished." After these jobs are run, the program will use Structure Harvester (or "fastChooseK.py" if wrapping fastStructure) to infer the optimal value of "K". Finally, the program will create plots with the inferred clustering, one for each calculated value of "K". A "Thermodynamic Integration" test will be performed to infer the bestK if using MavericK.

Results

After a successful run, inside the directory you selected as "output directory" (let's call it "My_results" for the sake of the example) you will find the following:

• In the root of "My_results" you will find the "results files" outputted by the wrapped program. One file (directory, in the case of MavericK) for each replicate of "K".
• Under "My_results/bestK" you will find either the results of the "Evanno test", the results of "fastChooseK.py", or the results of "Thermodynamic Integration" test, depending on what program was wrapped.
• Under "My_results/plots" you will find one plot for each value of "K" in SVG format.
• If logging was turned on, you will also find a detailed log file for each run in the root of "My_results".

Obtaining STRUCTURE alpha values

In order to obtain the alpha values of each STRUCTURE run, you can pass the option --log 1 to Strucutre_threader. This will write a file named K<knum>_rep<rnum>.stlog, where knum and rnum are the K value and replicate number respectively. This file contains the STDOUT generated by STRUCTURE. These can be used to plot alpha vs. iteration plots to help assess MCMC chain convergence.