Given one protein sequence and a multiple sequence alignment(MSA) of a set of proteins, I want to align the protein sequence with that MSA with out changing the MSA. Do you know any tool that is cable of doing this?
Paulo is right: Clustal could always do this. Clustal2 and Clustal-Omega have something called profile alignment-mode, which allows you to align two pre-aligned sequence-sets (profiles) in one step. Your one sequence can also be seen as a profile. Clustal2 furthermore had a special mode, which allows you to align (unaligned) sequences one by one to a profile (see -sequences flag below). This is interesting if you have many unaligned sequences that you want to align to a profile.
Assuming your one sequence is called one.fa and the prealigned others are in prof1.fa then the command-line would look like this:
The -sequences flag would align every (unaligned) sequence in profile2 to profile1. In your case you have just one sequence, so the flag doesn't do much and you could also leave it away.
Clustal Omega only knows about profile-profile alignment, so there's no -sequence flag. But again, in your case a normal profile-profile alignment will do, as the one sequence will be treated as an alignment.
You can (or at least could - I haven't done it in a while) do this with MUSCLE
MUSCLE allows you to align two alignments with each other (known as profile-profile alignment), although this functionality is also only available from the command line, not e.g. via the EBI MUSCLE web server. If one of the profiles contains just a single sequence, then you can do what you've asked about in the question