Jargon

Let's talk about some of the "jargon" we use when talking about PASV.

MSA: multiple sequence alignment

I will use this when talking about alignment as done by Clustal Omega and MAFFT.

This is in contrast to using HMMs to align sequences.

HMM: hiddon markov model

pasv can also use HMMs to align query sequences

PASV vs pasv

When you see pasv written in that font, I am talking about the PASV command line application.

PASV in all caps is the acronym for protein active site validation. It's both talking about the whole pasv program pipeline as well as the general idea of validating aspects of your query sequences.

Queries & query sequences

These are the sequences you are interested in. Often these will be metagenomic sequences, or sequences output from homology search.

pasv will assign a signature to each query sequence based on the residues it has at the specified key positions.

References & reference sequences

These are the sequences you are using as "references".

Depending on the command you're running, they could be sequences to align with queries in a multiple sequence alignment, or a profile HMM used to align queries against.

Key reference sequences, positions & residues

The pasv program will check residues in a particular column of an alignment.

To tell pasv which alignment columns to check, you specify "key" residues with respect to a reference sequence. I call this reference, the "key reference sequence". It doesn't have to be anything special...it is simply the sequence with which the number will be with respect to.

The key positions are the positions you care about. They are numbered with respect to the key reference sequence.

For example, if you care about an active site at residue 200 in a protein from E. coli, then you would make that E. coli peptide your key reference sequence, and residue 200 would be your key position. Then, pasv will check the residues of the query sequences that align with residue 200 with respect to the key reference sequence.

Each pasv command deals with key references and positions a bit differently. For specific info, see the command documentation.

Signatures & signature files

Each query sequence is assigned a "signature" based on the residues that query has at the given key residue positions.

A signature file lists these signatures and, optionally, whether a sequence spans a given region of interest (ROI).

The signature file is a TSV. Here is an example.

name pos_50 pos_52 pos_54 signature spans_start spans_end spans
AMK99662_spans_start_20_60_IQK I Q K IQK Yes No Start
AMK99662_spans_start_19_60_IQK I Q K IQK Yes No Start
AMK99662_21_60_IQK I Q K IQK No No Neither
AMK99662_spans_end_40_80_IQK I Q K IQK No Yes End
AMK99662_spans_end_40_81_IQK I Q K IQK No Yes End
AMK99662_40_79_IQK I Q K IQK No No Neither
AMK99662_spans_both_20_80_IQK I Q K IQK Yes Yes Both
AMK99662_spans_both_19_81_IQK I Q K IQK Yes Yes Both
AMK99662_21_79_IQK I Q K IQK No No Neither
AMK99662_full_length_IQK I Q K IQK Yes Yes Both
AMK99662_real_seq_IQK I Q K IQK Yes Yes Both
AMK99662_full_length_extra_IQK I Q K IQK Yes Yes Both
AMK99662_spans_start_20_60_ABC A B C ABC Yes No Start
AMK99662_spans_start_19_60_ABC A B C ABC Yes No Start
AMK99662_21_60_ABC A B C ABC No No Neither
AMK99662_spans_end_40_80_ABC A B C ABC No Yes End
AMK99662_spans_end_40_81_ABC A B C ABC No Yes End
AMK99662_40_79_ABC A B C ABC No No Neither
AMK99662_spans_both_20_80_ABC A B C ABC Yes Yes Both
AMK99662_spans_both_19_81_ABC A B C ABC Yes Yes Both
AMK99662_21_79_ABC A B C ABC No No Neither
AMK99662_full_length_ABC A B C ABC Yes Yes Both
AMK99662_real_seq_ABC A B C ABC Yes Yes Both
AMK99662_full_length_extra_ABC A B C ABC Yes Yes Both


  • This run checked for a region of interest as well as the key positions 50, 52, and 54.
  • spans_start: can be one of Yes, No, or NA.
  • spans_end: can be one of Yes, No, or NA.
  • spans: will be one of Neither, Start, End, or Both based on the values in spans_start and spans_end

Region of interest

The pasv program can optionally check that query sequences span a given region of interest.

You provide the start and end positions with respect to the key reference, and the program will check whether each query spans the start, spans the end, or spans both start and end.