Navigation

Suggested links

Bioinformatics at IMTECH

IMTECH at New Delhi

CSIR in INDIA

Resources

web-servers predicting Protein-Protein Interaction

Databases of Protein-Protein Interaction

Datasets of Protein-Protein Interaction

ProPrInt: Help page

Following are the detailed results of Support Vector Machine optimization

The result will help you to select optimum "threshold" for the prediction
[A] Escherichia coli
  (i)Amino Acid Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.00	0.565	1.000	0.968	1.000	0.967	0.739
0.800	0.709	1.000	0.979	1.000	0.978	0.833
0.600	0.811	1.000	0.986	1.000	0.985	0.894
0.400	0.887	1.000	0.992	1.000	0.991	0.938
0.200	0.930	1.000	0.995	1.000	0.995	0.962
0.000	0.957	1.000	0.997	0.999	0.997	0.976
-0.200	0.979	1.000	0.998	0.999	0.998	0.988
-0.400	0.990	1.000	0.999	0.997	0.999	0.993
-0.600	0.994	0.999	0.999	0.994	1.000	0.994
-0.800	0.996	0.993	0.994	0.921	1.000	0.955
-1.000	1.000	0.803	0.817	0.284	1.000	0.477

  (ii)Dipeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.790	1.000	0.985	1.000	0.984	0.882
0.800	0.884	1.000	0.992	1.000	0.991	0.936
0.600	0.932	1.000	0.995	1.000	0.995	0.963
0.400	0.962	1.000	0.997	1.000	0.997	0.979
0.200	0.981	1.000	0.999	1.000	0.998	0.989
0.000	0.992	1.000	0.999	1.000	0.999	0.996
-0.200	0.994	1.000	1.000	1.000	1.000	0.997
-0.400	0.996	1.000	1.000	1.000	1.000	0.998
-0.600	0.997	1.000	1.000	1.000	1.000	0.999
-0.800	0.997	1.000	1.000	1.000	1.000	0.999
-1.000	0.997	0.864	0.874	0.365	1.000	0.561

  (iii)Biochemical classes tripeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.564	1.000	0.968	1.000	0.967	0.738
0.800	0.721	1.000	0.980	1.000	0.979	0.840
0.600	0.839	1.000	0.988	1.000	0.988	0.910
0.400	0.914	1.000	0.994	1.000	0.993	0.953
0.200	0.946	1.000	0.996	1.000	0.996	0.971
0.000	0.972	1.000	0.998	1.000	0.998	0.985
-0.200	0.985	1.000	0.999	1.000	0.999	0.992
-0.400	0.993	1.000	0.999	1.000	0.999	0.996
-0.600	0.996	1.000	1.000	1.000	1.000	0.998
-0.800	0.999	0.999	0.999	0.990	1.000	0.994
-1.000	1.000	0.680	0.703	0.196	1.000	0.365

[B] Saccharomyces cerevisiae
  (i)Amino Acid Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.263	0.971	0.617	0.900	0.568	0.331
0.800	0.351	0.950	0.650	0.875	0.594	0.375
0.600	0.442	0.912	0.677	0.833	0.620	0.400
0.400	0.527	0.855	0.691	0.784	0.644	0.405
0.200	0.609	0.786	0.698	0.740	0.668	0.402
0.000	0.691	0.702	0.697	0.699	0.694	0.393
-0.200	0.762	0.600	0.681	0.656	0.716	0.367
-0.400	0.824	0.486	0.655	0.616	0.734	0.329
-0.600	0.878	0.371	0.624	0.582	0.752	0.288
-0.800	0.920	0.264	0.592	0.556	0.768	0.245
-1.000	0.952	0.171	0.561	0.534	0.780	0.196

  (ii)Dipeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.282	0.985	0.633	0.949	0.578	0.375
0.800	0.388	0.972	0.680	0.932	0.614	0.444
0.600	0.489	0.948	0.718	0.903	0.650	0.491
0.400	0.578	0.916	0.747	0.874	0.685	0.525
0.200	0.660	0.860	0.760	0.825	0.717	0.531
0.000	0.728	0.786	0.757	0.773	0.743	0.515
-0.200	0.791	0.682	0.736	0.713	0.765	0.476
-0.400	0.850	0.550	0.700	0.654	0.785	0.419
-0.600	0.900	0.404	0.652	0.602	0.802	0.351
-0.800	0.941	0.256	0.599	0.559	0.813	0.271
-1.000	0.969	0.135	0.552	0.528	0.813	0.188

  (iii) Biochemical classes tripeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.287	0.983	0.635	0.943	0.579	0.375
0.800	0.385	0.968	0.676	0.923	0.611	0.434
0.600	0.485	0.941	0.713	0.892	0.646	0.479
0.400	0.575	0.908	0.742	0.862	0.681	0.512
0.200	0.656	0.852	0.754	0.816	0.713	0.519
0.000	0.728	0.781	0.755	0.769	0.742	0.510
-0.200	0.789	0.676	0.733	0.709	0.763	0.469
-0.400	0.846	0.550	0.698	0.653	0.782	0.416
-0.600	0.897	0.406	0.652	0.602	0.798	0.348
-0.800	0.936	0.262	0.599	0.559	0.803	0.268
-1.000	0.962	0.144	0.553	0.529	0.792	0.185

[B] Helicobacter pylori
  (i)Amino Acid Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.324	0.986	0.655	0.957	0.593	0.413
0.800	0.452	0.976	0.714	0.950	0.640	0.502
0.600	0.559	0.954	0.757	0.924	0.684	0.558
0.400	0.660	0.910	0.785	0.880	0.728	0.589
0.200	0.753	0.868	0.811	0.851	0.779	0.626
0.000	0.825	0.800	0.813	0.805	0.821	0.626
-0.200	0.884	0.711	0.798	0.754	0.860	0.604
-0.400	0.932	0.600	0.766	0.700	0.898	0.564
-0.600	0.958	0.476	0.717	0.646	0.919	0.496
-0.800	0.979	0.307	0.643	0.585	0.935	0.385
-1.000	0.991	0.178	0.584	0.547	0.952	0.290

  (ii)Dipeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.300	0.995	0.647	0.982	0.587	0.409
0.800	0.486	0.988	0.737	0.975	0.658	0.547
0.600	0.630	0.971	0.800	0.956	0.724	0.639
0.400	0.747	0.942	0.844	0.928	0.788	0.702
0.200	0.824	0.910	0.867	0.902	0.838	0.737
0.000	0.885	0.852	0.868	0.857	0.881	0.737
-0.200	0.929	0.771	0.850	0.802	0.915	0.708
-0.400	0.965	0.661	0.813	0.740	0.950	0.657
-0.600	0.983	0.508	0.745	0.666	0.967	0.557
-0.800	0.993	0.291	0.642	0.583	0.977	0.399
-1.000	0.997	0.121	0.559	0.531	0.973	0.244

  (iii) Biochemical classes tripeptide Composition

Threshold	Sensitivity	Specificity	Accuracy	PPV	NPV	MCC
1.000	0.285	0.995	0.640	0.983	0.582	0.398
0.800	0.461	0.986	0.724	0.971	0.647	0.526
0.600	0.630	0.970	0.800	0.954	0.724	0.637
0.400	0.743	0.944	0.843	0.930	0.786	0.701
0.200	0.818	0.904	0.861	0.895	0.833	0.725
0.000	0.877	0.846	0.862	0.851	0.873	0.724
-0.200	0.922	0.765	0.843	0.797	0.907	0.695
-0.400	0.955	0.645	0.800	0.729	0.934	0.631
-0.600	0.980	0.490	0.735	0.658	0.961	0.540
-0.800	0.988	0.291	0.639	0.582	0.959	0.388
-1.000	0.997	0.123	0.560	0.532	0.978	0.248

Following are the methods of feature extraction from amino acid sequence

(i) Amino Acid Composition
This is simply the percentage frequency of all the 20 natural amino acids in the protein sequence. For a single protein sequence the vector represents 20 features.
Ci = fi / L
where Ci is the composition of ith amino acid; fi is frequency of ith amino acid in the sequence; and L is the total number of amino acid residue in the sequence.

(ii) Dipeptide Composition

mamoon

Created by: BIC at IMTECH 2008 | webs.iiitd.edu.in/raghava/