<< README for the Additional file 3 of "Characterization of multiple sequence alignment errors using complete-likelihood score and position-shift map" by Kiyoshi Ezawa >>

This directory, "additional_file_3/" contains the results of our manual validation analysis of the prototype script to classify MSA errors into different types.

The directories and files contained are as follows:

* README.txt
	This file.

* TEST_Outputs/
	The directory contains the automatic test outputs of the prototype Perl script ("classify_msa_errors_via_mblks.2val.pl,"  which is available in the directory, "ANALYSES/MSA_Prcd_Errors/Contrast_Analyses/Classify_Errors/Erroneous_Retions/Validation/Scripts/" of Additional file 2).

	Specifically, 
	the sub-directory, "TEST_Outputs/Test_MAFFT/class_0001/lclid_000000n/," contains the results of errors on the n th MSA via MAFFT (E-INS-i).
	And the sub-directory, "TEST_Outputs/Test_PRANK/class_0001/lclid_000000n/," contains the results of errors on the n th MSA via Prank ("best-fit").
	Here, n = 1,2,3,4,5 for both methods.
	The MSAs were created by simulation (via Dawg) along the phylogenetic tree of 15 mammals.

	More specifically, in each of the above sub-directories,

	    the file, "tbl_summ_class_msa_errors.2.txt," tabulates the main outputs (including the error classifications) on the erroneous segments.

	    The files, "LogFiles/log_seg_anal.mm.txt" (mm = (0), 1, ....), record the detailed outputs of the prototype Perl script, including the position-shift maps, their partitioning into blocks, and the association of errors with blocks.

	    And the file, "log_classify_msa_errors.txt," records some conditions under which the script was run,
		as well as some auxiliary outputs that will aid the interpretations of the automatic classifications,
		such as the position-shift map between the reconstructed and the intermediate MSAs,
		and that between the intermediate and true (i.e., reference) MSAs,
		when an attempt was made to associate a pair of blocks with a pair of errors.


* Validation/
	The directory contains the results of the manual validation, based on the outputs in "TEST_Outputs/." 

	Specifically,
	the file, "manual_validation_of_classification.xls," tabulates the results of our manual validation of the prototype Perl script. (More specifically, the file shows the comparison of our manual classification of each error and its automated classification via the prototype Perl script.

	And the file, "METHODS_RESULTS.txt," describes the methods and summalizes the results of the manual validation.


# This file was written by Kiyoshi Ezawa on Tuesday, January 12th, 2016.

