Biological Feature Selection and Disease Gene Identification using New Stepwise Random Forests

다국어 초록 (Multilingual Abstract)

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and diseasegene identification.

참고문헌 (Reference)

1 Kohavi, R., "Wrappers for feature selection" 97 (97): 273-324, 1997

2 Köhler, S., "Walking the interactome for prioritization of candidate disease genes" 82 : 949-958, 2008

3 Liu, Y., "Variable selection via a combination of the L0 and L1 penalties" 16 : 782-798, 2007

4 Finn, R., "The pfam protein families database" 38 : 211-222, 2010

5 Oti, M., "The modular nature of genetic diseases" 71 : 1-11, 2007

6 Goh, K., "The human disease network" 104 : 8685-8690, 2007

7 Gene Ontology Consortium, "The gene ontology database and informatics resource" 32 (32): 258-261, 2004

8 Kenji, K., "The feature selection problem: traditional methods and a new algorithm" 129-134, 1992

9 Hastie, T., "The elements of statistical learning" Springer 2001

10 Jeong, J. W., "The P160 steroid receptor coactivator 2, SRC-2, regulates murine endometrial function and regulates progesterone-independent and -dependent gene expression" 148 : 4238-4250, 2007

11 Yang, H., "The DNase domaincontaining protein TATDN1 plays an important role in chromosomal segregation and cell cycle progression during zebrafish eye development" 11 (11): 4626-4632, 2012

12 Erol, A., "Systemic DNA damage response and metabolic syndrome as a premalignant state" 10 (10): 321-334, 2010

13 Fan, J., "Sure independence screening for ultrahigh dimensional feature space" 70 : 849-911, 2008

14 Adie, E., "Speeding disease gene discovery by sequence based candidate prioritization" 6-55, 2005

15 Cao, H., "Single-nucleotide polymorphisms of the proprotein convertase subtilisin/kexin type 5 (PCSK5) gene" 46 : 730-732, 2001

16 Blum, A., "Selection of relevant features and examples in machine learning" 97 (97): 245-271, 1997

17 Bollmann, P., "Restricted Evaluation in Information Retrieval" 1981

18 Tong, L., "Recent advances: Transglutaminase in ocular health and pathological processes" 2011

19 Breiman, L., "Random forests" 45 : 5-32, 2001

20 Zhong, T., "RING finger ubiquitin-protein isopeptide ligase Nrdp1/FLRF regulates parkin stability and activity" 280 (280): 9425-9430, 2005

21 Ideker, T., "Protein networks in disease" 18 : 644-652, 2008

22 Yang, P., "Positive-unlabeled learning for disease gene identification" 28 (28): 2640-2647, 2012

23 Martimbeau, S., "Physiological cell death in endocrine-dependent tissues: An ovarian perspective" 46 (46): 241-254, 1997

24 Ring, H. G., "Pancreatic carcinoma with metastasis to the optic nerve" 77 (77): 798-800, 1967

25 Uttara, B., "Oxidative stress and neurodegenerative diseases:A review of upstream and downstream antioxidant therapeutic options" 7 (7): 65-74, 2009

26 Brown, K., "Online predicted human interaction database" 21 (21): 2076-2082, 2005

27 Chen, Y., "Mutation of NIMArelated kinase 1 (NEK1) leads to chromosome instability" 2011

28 Kondo, M., "Molecular cloning of human TAK1and its mutational analysis in human lung cancer" 75 (75): 559-563, 1998

29 McKusick, V., "Mendelian inheritance in man and its online version, OMIM" 80 : 588-604, 2007

30 Mitchell, M., "Machine learning" WCB 1997

31 Johansson, T., "Lack of nuclear expression of hairy and enhancer of split-1 (HES1) in pancreatic endocrine tumors" 40 (40): 354-359, 2008

32 Yang, P., "Inferring gene-phenotype associations via global protein complex network propagation" 6 (6): e21502-, 2011

33 Prasad, T., "Human protein reference database" 37 : 767-772, 2009

34 Smalter, A., "Human disease-gene classification with integrative sequencebased and topological features of protein-protein interaction networks"

35 Olmez-Hanci, T., "H2O2/UV-C oxidation of potential endocrine disrupting compounds: A case study with dimethyl phthalate" 8 (8): 620-627, 2009

36 Tsujie, M., "Growth inhibition of pancreatic cancer cells through activation of peroxisome proliferator-activated receptor Gamma/Retinoid X Receptor Alpha pathway" 23 (23): 325-331, 2003

37 Dufour, C. R., "Genome-wide orchestration of cardiac functions by the orphan nuclear receptors ERR alpha and Gamma" 5 (5): 345-356, 2007

38 Zhang, H., "Gene selection using support vector machines with nonconvex penalty" 22 : 88-95, 2006

39 Aerts, S., "Gene prioritization through genomic data fusion" 24 : 537-544, 2006

40 Xin, G., "GRIN1 regulates μ-opioid receptor activities by tethering the receptor and G protein in the lipid raft" 284 (284): 36521-36534, 2009

41 Tew, K., "Functional centrality:Detecting lethality of proteins in protein interaction networks" 2007

42 Brunner, H., "From syndrome families to functional genomics" 5 : 545-551, 2004

43 Suk, S., "Formation of nuclear isopeptide in the process of neuronal cell death following interstitial hyperthermia in normal rat brain" 19 (19): 633-640, 2001

44 Botta, V., "Exploiting SNP Correlations within Random Forest for Genome-Wide Association Studies" 2014

45 Day, B. W., "EphA3maintains tumorigenicity and is a therapeutic target in glioblastoma multiforme" 23 (23): 238-248, 2013

46 Yang, P., "Ensemble positive unlabeled learning for disease gene identification" 9 (9): 2014

47 Flicek, P., "Ensembl" 39 (39): 800-806, 2011

48 Giallourakis, C., "Disease gene discovery though integrative genomics" 6 : 381-406, 2005

49 Botstein, D., "Discovering genotypes underlying human phenotypes: Past successes for mendelian disease, future approaches for complex disease" 33 : 228-237, 2013

50 Xu, J., "Discovering disease-genes by topological features in human protein-protein interaction network" 22 (22): 2800-2805, 2006

51 Qiu, Y., "Detecting disease associated modules and prioritizing active genes based on high throughput data" 11-26, 2010

52 Guertin, D. A., "Defining the role of mTOR in cancer" 12 (12): 9-22, 2007

53 Karanika, S., "DNA damage response and prostate cancer: Defects, regulation and therapeutic implications" 2014

54 Katz, A. M., "Cytoskeletal abnormalities in the failing heart out on a LIM?" 101 (101): 2672-2673, 2000

55 Hall, M., "Correlation-based Feature Selection for Machine Learning" 1999

56 Breiman, L., "Classification and Regression Trees" CRC Press 1991

57 Chiong, M., "Cardiomyocyte death:mechanisms and translational implications" 2 : e244-, 2011

58 Mahaffey, K. R., "Blood organic mercury and dietary mercury intake: National health and nutrition examination survey" 112 (112): 562-570, 2004

59 Hultsch, C., "Biodistribution and catabolism of 18F-labelled isopeptide N(epsilon)-(gamma-glutamyl)-L-lysine" 29 (29): 405-413, 2005

60 Deshmukh, M., "Bax-inhibiting peptide protects cells from polyglutamine toxicity caused by Ku70 acetylation" 14 : 2058-2067, 2007

61 Szende, B., "Antitumor effect of lysine-isopeptides" 2002

62 Usmani-Brown, S., "Analysis of β-cell death in type 1 diabetes by droplet digital PCR" 155 (155): 3694-3698, 2014

63 Renk, G., "Analogue pigment studies of chromophore-protein interactions in metarhodopsins" 28 (28): 907-912, 1989

64 Guyon, I., "An introduction to variable and feature selection" 3 : 1157-1182, 2003

65 Radivojac, P., "An integrated approach to inferring gene-disease associations in humans" 72 (72): 1030-1037, 2008

66 Zou, H., "An improved 1-norm support vector machine for simultaneous classification and variable selection" 2 : 675-681, 2007

67 Johnson, V. J., "Aluminum-maltolate induces apoptosis and necrosis in neuro-2a cells: Potential role for p53 signaling" 83 (83): 329-339, 2005

68 Neil, J. R., "Altered TAB1: I KappaB kinase interaction promotes transforming growth factor beta-mediated nuclear factor-kappaB activation during breast cancer progression" 68 (68): 1462-1470, 2008

69 Jiang, R., "A random forest approach to the detection of epistatic interactions in case-control studies" 2009

70 Wang, Z. D., "A new method to measure the semantic similarity of GO terms" 23 (23): 1274-1281, 2007

71 Wang, G., "A forestbased feature screening approach for large-scale genome data with complex structures" 2015

72 Greenwood, P., "A Guide to Chisquared Testing" John Wiley & Sons 1996

73 Zhu, J., "1-norm support vector machines" 2004