Protein attributes contribute to halo-stability, bioinformatics approach

Ebrahimie, Esmaeil; Ebrahimi, Mansour; Sarvestani, Narjes Rahpayma; Ebrahimi, Mahdi

doi:10.1186/1746-1448-7-1

Aquatic Biosystems

Table 1 Results of supervised feature selection on the first 100 important protein attributes (with value equal to 1) contributing to halo-stability of studied proteins.

From: Protein attributes contribute to halo-stability, bioinformatics approach

No	Field	Value	Rank	No	Field	Value	Rank
1	Freq of Leu-Leu	1.0	Important	51	Freq of Tyr-Leu	1.0	Important
2	Freq of Phe-Pro	1.0	Important	52	Freq of Glu	1.0	Important
3	Freq of Leu-Ile	1.0	Important	53	Freq of Tyr-Ile	1.0	Important
4	Freq of Glu-Met	1.0	Important	54	Freq of carbon	1.0	Important
5	Freq of Trp-Lys	1.0	Important	55	Freq of Ile-Ile	1.0	Important
6	Freq of Leu	1.0	Important	56	Freq of Ile-Ala	1.0	Important
7	Freq of Val-Phe	1.0	Important	57	Freq of Met	1.0	Important
8	Freq of Trp-Pro	1.0	Important	58	Freq of Other	1.0	Important
9	Freq of Lys-Leu	1.0	Important	59	Freq of Val	1.0	Important
10	Freq of Leu-His	1.0	Important	60	Freq of Ala-Lys	1.0	Important
11	Freq of Leu-Arg	1.0	Important	61	Freq of Arg-Ile	1.0	Important
12	Freq of Ile	1.0	Important	62	Freq of Ile-Gly	1.0	Important
13	Freq of Val-Lys	1.0	Important	63	Freq of Lys-Thr	1.0	Important
14	Freq of Pro-Tyr	1.0	Important	64	Freq of Lys-His	1.0	Important
15	Freq of Val-Leu	1.0	Important	65	Freq of Phe-Ile	1.0	Important
16	Freq of Ser-Phe	1.0	Important	66	Freq of sulfur	1.0	Important
17	Freq of Val-Gln	1.0	Important	67	Freq of Ser-His	1.0	Important
18	Freq of Phe-Leu	1.0	Important	68	Freq of Lys-Val	1.0	Important
19	Freq of Asp-Trp	1.0	Important	69	Freq of Leu-Ser	1.0	Important
20	Freq of Gly-Leu	1.0	Important	70	Freq of His-Ser	1.0	Important
21	Freq of Leu-Trp	1.0	Important	71	Freq of Ala-Phe	1.0	Important
22	Freq of His	1.0	Important	72	Freq of nitrogen	1.0	Important
23	Freq of Phe	1.0	Important	73	Freq of Glu-Ser	1.0	Important
24	Freq of Lys	1.0	Important	74	Freq of Arg	1.0	Important
25	Freq of Tyr-Trp	1.0	Important	75	Freq of Met-Gly	1.0	Important
26	Freq of Gln-Leu	1.0	Important	76	Freq of Ile-Thr	1.0	Important
27	Freq of Leu-Val	1.0	Important	77	Freq of Pro-Leu	1.0	Important
28	Freq of Cys-Tyr	1.0	Important	78	Freq of Lys-Ile	1.0	Important
29	Freq of Leu-Lys	1.0	Important	79	Freq of Try	1.0	Important
30	Freq of Met-Leu	1.0	Important	80	Freq of Phe-Thr	1.0	Important
31	Freq of Thr-Phe	1.0	Important	81	Freq of Leu-Pro	1.0	Important
32	Freq of Val-Ile	1.0	Important	82	Freq of Ile-Val	1.0	Important
33	Freq of Leu-Gly	1.0	Important	83	Freq of Ile-Trp	1.0	Important
34	Freq of Gly-Ile	1.0	Important	84	Count of Trp-Lys	1.0	Important
35	Freq of Hydrophobic	1.0	Important	85	Freq of His-Leu	1.0	Important
36	Freq of Tyr-Lys	1.0	Important	86	Freq of Gly-Ala	1.0	Important
37	Freq of Thr-Val	1.0	Important	87	Freq of Ala-Val	1.0	Important
38	Freq of Ile-Asn	1.0	Important	88	Count of Trp-Pro	1.0	Important
39	Freq of His-Gln	1.0	Important	89	Freq of Val-Pro	1.0	Important
40	Freq of Glu-Gly	1.0	Important	90	Freq of Ser-Ile	1.0	Important
41	Freq of Leu-Tyr	1.0	Important	91	Freq of Glu-Lys	1.0	Important
42	Freq of Met-Arg	1.0	Important	92	Freq of oxygen	1.0	Important
43	Freq of Ala-Leu	1.0	Important	93	Freq of Thr-Leu	1.0	Important
44	Freq of Gln-Met	1.0	Important	94	Freq of Leu-Cys	1.0	Important
45	Freq of Trp-Leu	1.0	Important	95	Freq of Ile-Leu	1.0	Important
46	Freq of Thr-His	1.0	Important	96	Freq of Leu-Ala	1.0	Important
47	Freq of Ile-Arg	1.0	Important	97	Freq of Phe-Val	1.0	Important
48	Freq of Pro-Val	1.0	Important	98	Freq of Thr-Lys	1.0	Important
49	Freq of Tyr-Phe	1.0	Important	99	Freq of Tyr	1.0	Important
50	Count of hydrogen	1.0		100	Freq of Glu-Ala	1.0	Important

The algorithm considers one attribute at a time to determine how well each predictor alone predicts the target variable. The important value for each variable is then cal-culated as (1-p), where p is the p value of the appropriate test of association between the candidate predictor and the target variable. Since the target value was categorical, p values based on the F statistic was used.

Back to article page