align.txt

> model1 (this line should start with '>')
TNQKTKELSNLIETFAEQSRVLEKECT-KIGSKRDSKELRYKIETELIPNCTSVRDKIESN-I-L
IH-----QNGKLSADFKNLKTKYQSLQQSYNQRKSLFPLK

> model2 (this line should start with '>')
--E--DPFQQVVKDTKEQLNRINNYITRHNT---AD-DQEEEIQD-ILKDVEETIVDLDRSIIVM
K-RDENEDVSGREAQVKNIKQQLDALKLRFDRR--I----

>>>> the following lines (starting from third '>') will be ignored by TM-align: <<<<<

1, this is an example of fasta-format file to specify seed alignment used
   by TM-align structure alignment program.

2, This file should include alignments for two protein sequences. The 
   alignment entry can have any width (but should be <5,000).

3, Each alignment should start from the first amino acid of the target
   sequences. The ending unaligned residure can be ignored.

4, There should be at least three pairs of aligned residues, because
   superposition for <3 pairs of atoms is not mathematically defined.

5, More explanation on FASTA format can be seen at http://zhanglab.ccmb.med.umich.edu/FASTA