FProm

Human promoter prediction

Method description:

Program predicts potential transcription start positions by linear discriminant function combining characteristics describing functional motifs and oligonucleotide composition of these sites. FProm uses file with transcription factor binding sites from currently supported functional site data base.

For approximately 50-55% level of true promoter region recognition, FProm program will give one false positive prediction for about 4000 bp.

Another promoter recognition program, TSSG, uses promoter.dat file with selected factor binding sites (TFD, Ghosh,1993).

Prediction accuracy for each promoter type
Promoter Type A: TATA-less promoter
Sensitivity Specificity Threshold* Length**
1.000000 0.198215 -9.496 1.32975
0.990000 0.646996 -6.025 3.02029
0.950000 0.917724 -2.414 12.9585
0.900000 0.968909 +0.0467 34.2921
0.800000 0.992493 +3.329 142.028
0.700000 0.997591 +5.342 442.657
0.600000 0.998801 +6.508 889.255
0.500000 0.999409 +7.621 1805.3
0.400000 0.999705 +8.596 3610.59
0.300000 0.999858 +9.598 7491.98
0.200000 0.999911 +10.66 11987.2
0.100000 0.999968 +12.14 33297.7
Promoter Type B: TATA promoter
Sensitivity Specificity Threshold* Length**
1.000000 0.773441 -6.766 71.1151
0.990000 0.965914 -2.318 472.68
0.950000 0.996183 +1.117 4220.83
0.900000 0.998333 +2.528 9667.06
0.800000 0.999570 +4.613 37459.9
0.700000 0.999785 +6.41 74919.8
0.600000 0.999839 +7.963 99893
0.500000 0.999946 +9.586 299679
0.400000 0.999946 +11.21 299679
0.300000 0.999946 +12.5 299679
0.200000 1.000000 +14.14 1e+06
0.100000 1.000000 +16.54 1e+06

*Threshold value used by the program for a giver level of sensitivity
**Average length which contains 1 false-positive promoter.

References:
1. Solovyev V.V., Salamov A.A. (1997)
The Gene-Finder computer tools for analysis of human and model organisms genome sequences.
In Proceedings of the Fifth International Conference on Intelligent Systems for Molecular Biology (eds.Rawling C.,Clark D., Altman R.,Hunter L.,Lengauer T.,Wodak S.), Halkidiki, Greece, AAAI Press,294-302.

2. Solovyev V.V. (2001)
Statistical approaches in Eukaryotic gene prediction.
In Handbook of Statistical genetics (eds. Balding D. et al.), John Wiley & Sons, Ltd., p. 83-127.

3. Solovyev VV, Shahmuradov IA. (2003)
PromH: Promoters identification using orthologous genomic sequences.
Nucleic Acids Res. 31(13):3540-3545.

FProm output: 


Sequence    1 of    1,  Name: Homo sapiens chromosome 21; range 31946321 - 31958321; length 12001
Length of sequence:     12001
     7 promoter/enhancer(s) are predicted
Promoter Pos:      6473 LDF:   +8.734 
Promoter Pos:      3102 LDF:   +5.824 
Promoter Pos:      6078 LDF:  +16.297 TATA box at       6049    +5.597 TATAAAGT Enchancer at:      5942 Score:  +12.499
Promoter Pos:      1363 LDF:   +5.235 TATA box at       1336    +6.514 AATAAAAG 
Promoter Pos:      7068 LDF:   +1.165 TATA box at       7039    +4.190 TAAAAATA 
Promoter Pos:      9650 LDF:   +1.051 TATA box at       9618    +4.491 GTTAAAAA 
Promoter Pos:      5541 LDF:   +0.455 TATA box at       5512    +7.353 TATAAAAA

where

7 promoter/enhancer(s) are predicted Number of predicted promoters in this sequence.
Each line below defines an appropriate predicted promoter. Detailed description of a line from this list is shown further:

6078 LDF:  +16.297 TATA box at       6049    +5.597 TATAAAGT Enchancer at:      5942 Score:  +12.499
Promoter Pos: 6078 Position of TSS on DNA.
LDF: +16.297 value of Fisher's linear discriminant for the current promoter. A bigger value corresponds to more reliable promoter.
If a promoter belongs to class of TATA-containing promoters, the following fields are added:
TATA box at 6049 TATA-box position in the current promoter
+5.597 Score of this TATA-box
TATAAAGT Nucleotide sequence of this TATA-box
If there is an enhancer in proximity to the current promoter, the following fields are added:
Enchancer at: 5942 The position of enhancer in this promoter
Score: +12.499 Score of this enchancer