Results from AQUA

Extraction of data from RDB file 'AquaLacc_xtst.rdb'


RDB header

# Perl-RDB
# 
# PROGRAM        AQUA 
# DATE           file generated on Wed May 1 0:24:55 CET 1996
#  
# RESULTS FOR LIST: RANK vs. ACCURACY/COVERAGE
#  
# OVERALL SCORES:
# nPairs         4
# nPred          4
# nTrue          2
# rankFst        1
# rankAcc        50
# rankCov        100
# rankFaTb       2
# zHisto         3 ,	2
# zNocc          4 ,	4
# zNcor          2 ,	2
# zAcc           50 ,	50
# zCov           100 ,	100
#  
# NOTATION OVERALL SCORES:
# nPairs         number of pairs in predicted alignment
# nPred          number of predicted pairs included in analysis
#                (i.e. not found in Exclude set, zscore, and length ok asf.)
# nTrue          number of pairs in true alignment that were not excluded
#                (i.e. found in Search set NOT found in Exclude set, zscore ok asf.)
# rankFst        rank of first correct hit
# rankAcc        100 * (number of correct hits/number of hits)
#                note: appropriately changed if Exclude and Search set given!
# rankCov        100 * (number of correct hits/number of true hits)
#                note: appropriately changed if Exclude and Search set given!
# rank50         rank at which number of correct hits equals number of false hits
# rankFaTb	 rank at which number of false positive hits above this rank (fA)
#                equals the number of correct hits below this rank (tB)
# zHisto         intervals for z-dependent rank analysis 
#                (take all pairs i with z(i) > zHisto)
# zNocc          number of pairs with z(i) > zHisto
# zNcor          number of correctly predicted pairs with z(i) > zHisto
# zAcc           z-dependent accuracy = 100 * (zNcor/zNocc)
# zCov           z-dependent coverage = 100 * (zNcor/[min(nTrue,zNocc)])
# shiftAcc       AQUAshift accuracy (see below) for first correct hit
# shiftCov       AQUAshift coverage (see below) for first correct hit
# shiftPide      AQUAshift percentage identity (see below) for first correct hit
#        
#  
# NOTATION RANK ANALYSIS:
# rankAqua       rank in analysis (skipping pairs that were excluded)
# listNcor       cumulative number of correctly predicted pairs at given rank
# ok             =1, hit was found in true alignment
# listAcc        accuracy = 100 * (listNcor/rankAqua)
# listCov        coverage = 100 * (listNcor/[min(nTrue,rankAqua)])
#         



RDB table

rankAqua ok listNcor listAcc listCov
11 1 100.0 100.0
20 1 50.0 50.0
30 1 33.3 50.0
41 2 50.0 100.0


Aqua Home Daf Format Mail to Rost