Reconstruct Peptide Sequences!

SPIDER is a sequence tag based homology search tool, which can be used to identify peptides that traditional tandem mass spectrometry ion search tools will miss. This is because it does not require an exact match to the spectrum, only similar to the proposed sequence. PEAKS auto de novo sequencing is an ideal way to generate them.

If the protein being analyzed is not in a database, SPIDER can identify peptides from the database that show significant homology to the de novo sequence. The ultimate need for this tool derives from using unsequenced organisms.

Essentially SPIDER treats both sequences, obtained from de novo sequencing and database searching, as independent candidates, each independently 'lying'. Rule of thumb: Assume each "liar" lies slightly and independently. Then the truth should be close to both lies.

It should be emphasized here that BLAST IS NOT a peptide homology search tool. Many researchers have misunderstood that, resulting in, not just poor, but wrong data analysis. SPIDER is tolerant of common de novo sequencing errors such as (I/L), (N/GG) and (SAT/TAS). Thus it will not penalize the alignment if certain residues or combinations of the same mass are substituted, where as BLAST would fail here.

SPIDER can also be used to improve coverage on a protein that is already in a database, by seeking exact sequence matches and being plugged into a specialized database.

The two methods available with SPIDER include tag match and homology match.

Tag Match: Rather than performing a dedicated mutation search, the tag match option insists that the mass of the peptide returned is the same as that of the de novo sequence. This search mode allows for the use of fixed modifications.

Homology Match: This is a more rigorous and a more resource intensive search mode, taking into account all types of mutations and positional confidence scores. This search mode also creates reconstructed peptides that use information from both the de novo sequence and the database sequence in order to characterize the real sequence. This search mode allows for the use of both fixed and variable modifications.

Peptide View

As noted, SPIDER will search the database for homologous peptides and attempt to consolidate these into protein hits as well. The result report will look very similar to the results for PEAKS database search and inChorus, with a few differences.
Within the Peptide View: 

• Resides the peptide details tab which displays SPIDER matches shown in red.
• Letters on a green background and with vertical bars, indicate agreement.
• Letters on a red background indicate sequencing error.
• Color codes on the de novo sequence letters still indicate positional confidence.
• Letters on a blue background indicate uncertainty or mutation.

When simply identifying exact peptides from the database, using PEAKS Protein ID, or inChorus, there's no need to reconstruct the 'real' sequence. This limits the amount of sequences required for processing and thus increases time efficiency.

Protein View

In addition to peptides, SPIDER offers a Protein View which yields a similar display to PEAKS Protein ID. This time, instead of homologous (blue) regions, there are now red regions to indicate areas of mutation found by SPIDER.

Sequence Browser

The specific sequence(s) noted in red are more clearly visible here in the Sequence Browser.

Related Reading:

1. Han Y, Ma B, Zhang K. SPIDER: Software for Protein Identification from Sequence Tags Containing De Novo Sequencing Error. J Bioinform Comput Biol. 2005 Jun;3(3):697-716.
2. Ma B, Yuen D. SPIDER: Novel Scoring Function Improves Homology Searches using MS/MS de novo Sequencing Results. ASMS 2008: ThP 648.
3. Yang W, Yuen D, Ma B, Rogers I. Improving Protein Coverage by de novo Sequence Homology Searching with SPIDER. ASMS 2007: MPK 176