Scfbio

Genome Analysis

Regulation of gene expression is a matter of chemistry between DNA and proteins at the molecular level. While remarkable advances have been made over the last two decades in gene identification using statistical and mathematical models with a heavy usage of databases and computational protocols, a gene finding model which directly captures the physicochemical properties intrinsic to DNA sequences and the chemistry of protein DNA interactions remains a goal yet to be realized.

Progressing towards development of an ab initio physico-chemical model christened ChemGenome, we constructed three dimensional vectors for each tri-nucleotide (codon) considering hydrogen bond energy, stacking energy and a third parameter which we provisionally identified with groove potentials. As this three-dimensional vector walks along a genome, the net orientation of the resultant vector differs significantly for gene and non-gene regions. The model works well for prokaryotic genomes and shows promise of universal applicability. Efforts to develop ChemGenome into a stand-alone algorithm for gene prediction are in progress. .

ChemGenome 1.1 Accuracy

S.No.	NCBI_ID	Species Name	Genes	TP#	FP#	SS#	SP#	CC#

1	NC_000117	Chlamydia trachomatis	463	458	4	0.98	0.99	0.98
2	NC_000853	Thermotoga maritima MSB8	641	619	3	0.96	0.99	0.96
3	NC_000854	Aeropyrum pernix K1	561	532	7	0.94	0.98	0.93
4	NC_000868	Pyrococcus abyssi GE5	632	630	241	0.99	0.63	0.49
5	NC_000907	Haemophilus influenzae	955	953	7	0.99	0.99	0.99
6	NC_000908	Mycoplasma genitalium G-37	189	186	2	0.98	0.98	0.97
7	NC_000909	Methanocaldococcus janaschii	720	708	9	0.98	0.98	0.97
8	NC_000912	Mycoplasma pneumoniae M129	243	241	2	0.99	0.99	0.98
9	NC_000913	Escherichia coli K12	2759	175	659	0.63	0.72	0.39
10	NC_000915	Helicobacter pylori	731	727	4	0.99	0.99	0.98
11	NC_000916	Methanobacterium thermoautotrophicum	719	711	4	0.98	0.99	0.98
12	NC_000917	Archaeoglobus fulgidus	782	774	8	0.98	0.98	0.97
13	NC_000917	Archaeoglobus fulgidus DSM4304	782	774	8	0.98	0.98	0.98
14	NC_000918	Aquifex aeolicus VF5	584	575	3	0.98	0.99	0.97
15	NC_000921	Helicobacter pylori strain J99	658	648	9	0.98	0.98	0.97
16	NC_000922	Chlamydophila pneumoniae CWL029	597	590	9	0.98	0.98	0.97
17	NC_000948	Borrelia burgdorferi B31 plsmids cp32-1	11	11	0	1.0	1.0	1.0
18	NC_000949	Borrelia burgdorferi B31 plsmids cp32-3	11	11	0	1.0	1.0	1.0
19	NC_000950	Borrelia burgdorferi B31 plsmids cp32-4	11	11	0	1.0	1.0	1.0
20	NC_000951	Borrelia burgdorferi B31 plsmids cp32-6	10	10	0	1.0	1.0	1.0

# True positives (TP): Genes evaluated as genes.
# False positives (FP): Non-genes evaluated as genes.
# True negatives (TN): Non-genes evaluated as non-genes.
# False negatives (FN): Genes evaluated as non-genes. Number of actual positives
(AP) = TP+FN. Number of actual negatives (AN) = FP+TN.
Predicted number of positives (PP) =TP+FP.
Predicted number of negatives (PN) = TN+FN. Sensitivity (SS) =TP / (TP+FN).
Specificity (SP) =TP / (TP+FP).

REFERENCE :

1.) Dutta, S., Singhal, P., Agrawal, P., Tomer, R., Kritee, Khurana, E. and Jayaram, B. A Physico-Chemical Model for Analyzing DNA sequences, J. Chem. Inf. Model, 2005. [ABSTRACT]

2.)Jayaram, B. Beyond the wobble: the rule of conjugates. J. Mol. Evol. 1997, 45, 704-705. [ABSTRACT]