Purpose: Given an alignment or set of unaligned nucleotide or protein sequences, this tool compares the sequences and eliminates any duplicates or very similar sequences, thus producing a set of unique sequences.
Details: By default, the program removes all non-letter characters from the sequences, converts all letters to uppercase, and considers as a "duplicate" any sequence that is a subsequence of a longer sequence (e.g., the sequence ATG is a duplicate of the sequence CATGCC). These three default behaviors can be modified by changing the first three options shown below. In the fourth option, you can choose to restore any gaps or non-uppercase characters that were present in the input. In the fifth option, you can decide how similar among sequences would be removed. Checking the box of sequences are aligned will dramatically speed up the program. Otherwise, it would run in the background with results emailed to you. The final option gives a means of automatically analyzing your input sequences as a series of sequence groups. The results page summarizes the duplicate and unique sequence sets and allows you to view and download the resulting unique sequences file and the duplicate sequences file.
Note: This program needs an alignment. If your sequences are NOT aligned, please uncheck the box at the bottom of the Input block. Adding an alignment step slows the program down dramatically.
For more details, see ElimDupes Explanation.