Home   -  Secondary structure  Comparative modelling  Contacts  Threading   - search   - FTP   - Mirrors    -   Bottom Explain scores

EVA sec results for: SSpro4

Organisation of method specific results

METHOD about: ../../doc/explain_methods.html#PX_about_sspro4
METHOD url: SSpro4
METHOD quote: G Pollastri, D Przybylski, B Rost, P Baldi:: Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles. Proteins, 47, 228-335, 2002
TYPE : sec
NPROT : 144
AVE_Q3 : 77.9
AVE_SOV: 72.0
NOTE 1 : here ALL results obtained for sspro4 are given, i.e. NOT those for a common subset!
NOTE 2 : all proteins evaluated had no SIGNIFICANT sequence identity to any protein of previously known structure!

Averages over all proteins
type date sots id Q3 Q3H%o Q3H%p Q3E%o Q3E%p Q3L%o Q3L%p sov sovH sovE sovL BAD info infoO infoP corrH corrE corrL class contDH contDE

Averages over each month
TypeMethodMonthNprotAverage Q3 for monthFile with details for month
secsspro4ALL 144 77.9  
secsspro42004_02 4 80.8 old/2004_02_sspro4.html
secsspro42004_03 13 75.3 old/2004_03_sspro4.html
secsspro42004_04 11 82.0 old/2004_04_sspro4.html
secsspro42004_05 6 72.0 old/2004_05_sspro4.html
secsspro42004_06 7 78.9 old/2004_06_sspro4.html
secsspro42004_07 10 80.0 old/2004_07_sspro4.html
secsspro42004_08 17 82.4 old/2004_08_sspro4.html
secsspro42004_09 4 71.2 old/2004_09_sspro4.html
secsspro42004_10 1 72.0 old/2004_10_sspro4.html
secsspro42004_11 4 77.1 old/2004_11_sspro4.html
secsspro42004_12 1 82.0 old/2004_12_sspro4.html
secsspro42005_02 1 46.8 old/2005_02_sspro4.html
secsspro42005_03 18 77.9 old/2005_03_sspro4.html
secsspro42005_04 12 78.8 old/2005_04_sspro4.html
secsspro42005_05 12 76.4 old/2005_05_sspro4.html
secsspro42005_06 4 76.8 old/2005_06_sspro4.html
secsspro42005_07 15 76.2 old/2005_07_sspro4.html
secsspro42005_08 16 78.5 old/2005_08_sspro4.html
secsspro42006_03 3 77.6 old/2006_03_sspro4.html
secsspro4ALL 144 77.9  

Notations used (detailed explanation)

psum  percentage over all proteins
sig  standard deviation
errsig significant difference (to distinguish between two methods)
type   prediction type
date   date when added [yyyy_mm_dd]
sots   standard of truth
id   identifier of protein
Q3   three-state per-residue accuracy (percentage of correctly predicted residues)
Q3H%o   correctly predicted helix residues (percentage of helix observed)
Q3H%p   correctly predicted helix residues (percentage of helix predicted)
Q3E%o   correctly predicted strand residues (percentage of helix observed)
Q3E%p   correctly predicted strand residues (percentage of helix predicted)
Q3L%o   correctly predicted loop residues (percentage of helix observed)
Q3L%p   correctly predicted loop residues (percentage of helix predicted)
sov   per segment accuracy (three states)
sovH   per segment accuracy for helix residues
sovE   per segment accuracy for strand residues
sovL   per segment accuracy for non-regular residues
BAD   percentage of confused HELIX/STRAND residues (helix predicted as strand, strand predicted as helix)
infoO   information index accuracy (percentage of observed)
infoP   information index accuracy (percentage of predicted)
corrH   Matthews correlation index for helix
corrE   Matthews correlation index for strand
corrL   Matthews correlation index for loop
var   variance over all proteins (1/N-1)*( score(i) - )**2
sig   sigma (standard deviation) over all proteins (sqrt[var])
errsig   significant differences ( standard dev / sqrt(Nprot) )
Explain scores - Home   -   Secondary structure  Comparative modelling  Contacts  Threading   - search   - FTP   - Mirrors    -   Top