ConSurfDB vs. ConSurf: Difference between revisions

With default parameters, the ConSurf Server results have an average [[#Average Pairwise Distance]] (APD) of 1.00<ref name="APD">Tested with 20 arbitrarily selected proteins, mostly enzymes. Average of the average pairwise distance (APD) values: 1.00; range 0.82-1.42.</ref>, and an average of only a few "yellow" residues with insufficient data.<ref name="ISD">Tested with 20 arbitrarily selected proteins, mostly enzymes. Average number of amino acids with insufficient data ("yellow" in ConSurf): 3.5; range 0 to 16.</ref> For the examples below, it was necessary to [[#Limiting ConSurf Analysis to Proteins of a Single Function|customize the ConSurf Server job parameters]] in order to reveal conservation present in proteins with the same function as the query.

[[2vaa]] contains three chains. Here, (<scene name='39/399854/2vaa_consurf_halos_w274_y159/4'>restore initial scene, ConSurf Server default settings, APD 1.1</scene>)<ref name="apd1.1">Default ConSurf Server settings: UniRef90 database, excluding sequences with > 95% or < 35% identity with the query, MSA has 150 sequences sampled evenly from all unique sequence hits.</ref> ConSurf colors are applied only to the alpha chain (chain A), while the beta chain (chain B = &beta;-2 microglobulin) and the 8 amino acid peptide (chain P) are shown as gray backbone traces.  

A custom consurf job resulting in an APD of 0.51<ref name="apd0.51">Custom ConSurf Server settings for APD 0.51: UniRef90 database, excluding sequences with > 95% or '''< 50%''' identity with the query, MSA has 150 sequences sampled evenly from all unique sequence hits.</ref> (NOT SHOWN) had '''NO groove residues with conservation levels > 6'''. Trp274 was level 9.

By default, ConSurf Server excludes from the multiple sequence alignment sequences with >95% identity, or <35% identity with the query sequence. Changing those limits to >98% and <70% reduced the default APD of 1.1 to 0.31<ref name="apd0.31">Custom ConSurf Server settings for APD 0.31: UniRef90 database, excluding sequences with '''> 98% or < 70%''' identity with the query, MSA has 150 sequences sampled evenly from all unique sequence hits.</ref>. <scene name='39/399854/2vaa_apd_point31/3'>This result reveals high conservation of the following 4 key residues in the groove</scene> (yellow halos). With spin OFF, touch a residue to identify it.

<scene name='39/399854/2vaa_peptide_contacts/1'>Here are all the polar residues contacting the peptide</scene>. Use the '''POPUP BUTTON''' to see details! (This scene is easily obtained in [http://firstglance.jmol.org FirstGlance]: Tools tab, click Contacts, check Label Contacts, and [[Help:How to Insert a ConSurf Result Into a Proteopedia Green Link|made into a Green Link]].

Another custom ConSurf Server job<ref name="apd0.30">Custom ConSurf Server settings for APD 0.30: UniRef90 database, excluding sequences with > 95% or < 35% identity with the query, MSA has '''250''' sequences '''closest''' to the query.</ref> gave an '''APD of 0.30''', but levels for the above 4 groove residues were 7-8. These lower levels can be accounted for by the highest expectation value<ref name="evalue" /> in the MSA, which was 10 to the power -141. In contrast, for the job with APD 0.31, the highest expectation value was 10 to the power -84.

<font color="red">UPDATE IN PROGRESS:</FONT> [[User:Eric Martz|Eric Martz]] 15:02, 29 July 2024 (UTC)

In order to discover key functional residues, it is important to inspect multiple ConSurf Server jobs for highly conserved residues, including multiple jobs with [[#Average Pairwise Distance]]s in the range 0.25-0.5 using the [[#Limiting ConSurf Analysis to Proteins of a Single Function|above methods]]. Residues conserved in the broader family of more distantly related proteins with different functions will typically be revealed with default ConSurf Server settings, or even in the ConSurf'''DB''' result.

==References==