GEGE: Predicting Gene Essentiality with Graph Embeddings

Abstract

A gene is considered essential if its function is indispensable for the viability or reproductive success of a cell or an organism. Distinguishing essential genes from non-essential ones is a fundamental question in genetics, and it is key to understanding the minimal set of functional requirements of an organism. Knowledge of the set of essential genes is also crucial in drug discovery. Several reports in the literature show that the gene location in a protein-protein interaction network is correlated with the target gene’s essentiality. Here, we ask whether the node embeddings of a protein-protein interaction (PPI) network can help predict gene essentiality. Our results on predicting human gene essentiality show that node embeddings alone can achieve up to 88% AUC score, which is better than using topological features to characterize gene properties and other previous work’s results. We also show that, when combined with homology information across species, this performance reaches 89% AUC. Our work shows that node embeddings of a protein in the PPI network capture the network connectivity patterns of the proteins and improve the gene essentiality predictions.

Keywords

• Graph representations
• Node embeddings
• Gene essentiality
• Network topological features
• Protein-protein interaction network

GEGE: Çizge Gömülümleriyle Gen Esaslılığını Tahmin Etme

Abstract

İşlevi, bir hücrenin veya organizmanın hayatta kalabilmesi veya üreme başarısı için vazgeçilmez olan genler, esaslı genler olarak kabul edilir. Esaslı genleri esaslı olmayanlardan ayırt etmek, bir organizmanın minimum fonksiyonel gereksinimlerinin anlaşılabilmesi için genetikte kilit bir sorudur. Esaslı genler küme bilgisi, ilaç tasarlanmasında da çok önemlidir. Literatürdeki, bir protein-protein etkileşim ağındaki gen konumunun, gen esaslılığı ile ilişkili olduğunu göstermiştir. Burada, bir protein-protein etkileşimi (PPI) ağının düğüm yerleştirmelerinin gen gerekliliğini tahmin etmeye yardımcı olup olamayacağını soruyoruz. İnsan geninin esaslığını tahmin etme konusundaki sonuçlarımız, düğüm gömülümlerinin tek başına %88'e kadar AUC skoruna ulaşabileceğini göstermektedir. Bu skor, gen özelliklerini karakterize etmek için topolojik özellikleri kullanılan modellerin başarımından ve önceki çalışma sonuçlarından daha iyidir. Ayrıca, türler arası homoloji bilgisi ile birleştiğinde, bu performansın %89 AUC skoruna ulaştığını gösteriyoruz. Çalışmamız, PPI ağındaki bir proteinin düğüm gömülümlerinin, proteinlerin ağ bağlantı modellerini yakaladığını ve gen esaslılık tahminlerini geliştirdiğini gösteriyor.

Keywords

• Çizge gösterimleri
• Düğüm gömülümleri
• Gen esaslılığı
• Ağ topolojik özellikleri
• Protein-protein etkileşim ağı

References

1. [1] G. Rancati, J. Moffat, A. Typas, N. Pavelka, “Emerging and evolving concepts in gene essentiality”, Nature Reviews Genetics, vol. 19, no.1, pp. 34, 2018.
2. [2] M. Itaya, “An estimation of minimal genome size required for life”, FEBS Letters, vol. 362, no.3, pp. 257–60, 1995.
3. [3] A. R. Mushegian, E.V. Koonin, “A minimal gene set for cellular life derived by comparison of complete bacterial genomes”, Proceedings of the National Academy of Sciences, vol. 93, no.19, pp. 10268–73, 1996.
4. [4] E.V. Koonin, “How many genes can make a cell: the minimal-gene-set concept”, Annual Review of Genomics and Human Genetics, vol. 1, no. 1, pp. 99–116, 2000.
5. [5] M.Y. Galperin, E.V. Koonin, “Searching for drug targets in microbial genomes”, Current Opinion in Biotechnology, vol. 10, no. 6, pp. 571–78, 1999.
6. [6] A.F. Chalker, R.D. Lunsford, “Rational identification of new antibacterial drug targets that are essential for viability using a genomics-based approach”, Pharmacology & Therapeutics, vol. 95, no. 1, pp. 1–20, 2002.
7. [7] H. Farmer, N. McCabe, C.J. Lord, A.N. Tutt, D.A. Johnson, T.B. Richardson, et al. “Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy”, Nature, vol. 434, no. 7035, pp. 917, 2005.
8. [8] N.J. O’Neil, M.L. Bailey, P. Hieter, “Synthetic lethality and cancer”, Nature Reviews Genetics, vol. 18, pp. 10, pp. 613, 2017.

9. [9] A. Cho, N. Haruyama, A.B. Kulkarni, “Generation of transgenic mice”, Current Protocols in Cell Biology, vol. 42, no. 1, chapter. 19, unit. 11, 2009.
10. [10] G. Giaever, A.M. Chu, L. Ni, C. Connelly, L. Riles, S. V´eronneau, et al. “Functional profiling of the Saccharomyces cerevisiae genome”, Nature, vol. 418, no. 6896, pp. 387–91, 2002.
11. [11] J.M. Silva, K. Marran, J.S. Parker, J. Silva, M. Golding, M.R. Schlabach, et al. “Profiling essential genes in human mammary cells by multiplex RNAi screening”, Science, vol. 319, no. 5863, pp. 617–20, 2008.
12. [12] T. Wang, K. Birsoy, N.W. Hughes, K.M. Krupczak, Y. Post, J.J. Wei, et al. “Identification and characterization of essential genes in the human genome”, Science, vol. 350, no. 6264, pp. 1096–101, 2015.
13. [13] M.A. D'Elia, M.P. Pereira, E.D. Brown, “Are essential genes really essential?”, Trends in Microbiology, vol. 17, no. 10, pp. 433–8, 2009.
14. [14] L.W. Ning, H. Lin, H. Ding, J. Huang, N.N.M. Rao, F.B. Guo, “Predicting bacterial essential genes using only sequence composition information”, Genetics and Molecular Research: GMR, vol. 13, no. 2, pp. 4564–72, 2014.
15. [15] W.C. Wei, L.W. Ning, Y.N. Ye, F.B. Guo. “Geptop: A gene essentiality prediction tool for sequenced bacterial genomes based on orthology and phylogeny”, PloS One; 2013.
16. [16] F.B. Guo, C. Dong, H.L. Hua, S. Liu, H. Luo, H.W. Zhang, et al. “Accurate prediction of human essential genes using only nucleotide composition and association information”, Bioinformatics, 33 12:1758–64, 2017.
17. [17] J. Deng, L. Deng, S. Su, M. Zhang, X. Lin, L. Wei, et al. “Investigating the predictability of essential genes across distantly related organisms using an integrative approach”, Nucleic Acids Research, vol. 39. no. 3, pp. 795-807, 2011.
18. [18] L. Chen, Y.H. Zhang, S. Wang, Y. Zhang, T. Huang, Y.D. Cai, “Prediction and analysis of essential genes using the enrichments of gene ontology and KEGG pathways”, PloS One, vol. 12, no. 9, e0184129, 2017.
19. [19] H. Jeong, S.P. Mason, A.L. Barabasi, Z.N. Oltvai, “Lethality and centrality in protein networks”, Nature, vol. 411, no. 6833, pp. 41-2, 2001.
20. [20] M.W. Hahn, A.D. Kern, “Comparative genomics of centrality and essentiality in three eukaryotic protein-interaction networks”, Molecular Biology and Evolution, vol. 22, no. 4, pp. 803–6, 2004.
21. [21] N.N. Batada, L.D. Hurst, M. Tyers, “Evolutionary and physiological importance of hub proteins”, PLoS Computational Biology, vol. 2, no. 7, e88, 2006.
22. [22] E. Zotenko, J. Mestre, D.P. O’Leary, T.M. Przytycka, “Why do hubs in the yeast protein interaction network tend to be essential: reexamining the connection between the network topology and essentiality”, PLoS Computational Biology, vol. 4, no. 8, e1000140, 2008.
23. [23] Y.C. Hwang, C.C. Lin, J.Y. Chang, H. Mori, H. F. Juan, H.C. Huang, “Predicting essential genes based on network and sequence analysis”, Molecular BioSystems, vol. 5, no.12, pp. 1672–78, 2009.
24. [24] J. Wang, M. Li, H. Wang, Y. Pan, “Identification of essential proteins based on edge clustering coefficient”, IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB), vol. 9, no. 4, pp. 1070–80, 2012.
25. [25] M.L. Acencio, N. Lemke, “Towards the prediction of essential genes by integration of network topology, cellular localization and biological process information”, BMC Bioinformatics, vol. 10, no. 1, pp. 290, 2009.
26. [26] J. Cheng, W. Wu, Y. Zhang, X. Li, X. Jiang, G. Wei, et al. “A new computational strategy for predicting essential genes”, BMC Genomics, vol. 14, no. 910, 2013.
27. [27] M.C. Palumbo, A. Colosimo, A. Giuliani, L. Farina, “Functional essentiality from topology features in metabolic networks: a case study in yeast”, FEBS Letters, vol. 579, no. 21, pp. 4642-6, 2005.
28. [28] T. Can, "ProtRank: A better measure for protein essentiality," in Proceedings of the 3rd International Symposium on Health Informatics and Bioinformatics (HIBIT'08), Istanbul, May 2008. [29] L. Page, S. Brin, R. Motwani and T. Winograd, “The Pagerank Citation Ranking: Bringing Order to the Web,” Technical Report, Stanford University, Stanford, 1998.
29. [30] S. Coulomb, M. Bauer, D. Bernard, M.C. Marsolier-Kergoat, “Gene essentiality and the topology of protein interaction networks”, Proceedings of the Royal Society of London B: Biological Sciences, vol. 272, no. 1573, pp. 1721–1725, 2005.
30. [31] X. He, J. Zhang, “Why do hubs tend to be essential in protein networks?”, PLoS Genetics, vol. no. 6, e88, 2006.
31. [32] H. Yu, P.M. Kim, E. Sprecher, V. Trifonov, M. Gerstein, “The importance of bottlenecks in protein networks: correlation with gene essentiality and expression dynamics”, PLoS Computational Biology, vol. 3, no. 4, e59, 2007.
32. [33] M.P. Joy, A. Brock, D.E. Ingber, S. Huang, “High-betweenness proteins in the yeast protein interaction network”, BioMed Research International, vol. 2005, no. 2, pp. 96–103, 2005.
33. [34] M. McPherson, L. Smith-Lovin, J.M. Cook, “Birds of a feather: Homophily in social networks”, Annual Review of Sociology, vol. 27, 1, 415–44, 2001.
34. [35] F. Lorrain, H.C. White, “Structural equivalence of individuals in social networks”, The Journal of Mathematical Sociology, vol. 1, no. 1, pp. 49–80, 1971.
35. [36] B. Perozzi, R. Al-Rfou, S. Skiena, “DeepWalk: Online Learning of Social Representations”, KDD: Proceedings International Conference on Knowledge Discovery & Data Mining, pp. 701–10, 2014.
36. [37] A. Grover, J. Leskovec, “node2vec: Scalable Feature Learning for Networks”, KDD: Proceedings International Conference on Knowledge Discovery & Data Mining, pp.855–864, 2016.
37. [38] R. Andersen, F. Chung, K. Lang, “Local graph partitioning using PageRank vectors”, IEEE, pp. 475–86, 2006.
38. [39] F. Fouss, A. Pirotte, J.M. Renders, M. Saerens, “Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation”, IEEE Transactions on Knowledge and Data Engineering, vol. 19, no. 3, pp. 355–69, 2007.
39. [40] Y. Chen, D. Xu. “Understanding protein dispensability through machine-learning analysis of high-throughput data”, Bioinformatics, vol. 21, no. 5, pp. 575–81, 2004.
40. [41] J. Leskovec, R. Sosic, “SNAP: A general-purpose network analysis and graph-mining library”, ACM Transactions on Intelligent Systems and Technology (TIST), vol. 8, no. 1, pp. 1, 2016.
41. [42] T. Hart, M. Chandrashekhar, M. Aregger, Z. Steinhart, K.R. Brown, G. MacLeod, et al. “High-resolution CRISPR screens reveal fitness genes and genotype-specific cancer liabilities”, Cell, vol. 163, no. 6, pp. 1515–26, 2015.
42. [43] V.A. Blomen, P. Majek, L.T. Jae, J.W. Bigenzahn, J. Nieuwenhuis, J. Staring, et al. “Gene essentiality and synthetic lethality in haploid human cells”, Science, vol. 350, no. 6264, pp.1092–6. 2015.
43. [44] J.M. Silva, K. Marran, J.S. Parker, J. Silva, M. Golding, M.R. Schlabach, et al. “Profiling essential genes in human mammary cells by multiplex RNAi screening”, Science, vol. 319, no. 5863, pp. 617–20, 2008.
44. [45] R. Marcotte, K.R. Brown, F. Suarez, A. Sayad, K. Karamboulas, P.M. Krzyzanowski et al. “Essential gene profiles in breast, pancreatic, and ovarian cancer cells”, Cancer Discovery, vol. 2, no. 2, pp. 172–89, 2012.
45. [46] J. Luo, M.J. Emanuele, D. Li, C.J. Creighton, M.R. Schlabach, T. Westbrook, et al. “A Genome-wide RNAi screen identifies multiple synthetic lethal interactions with the Ras oncogene”, Cell, vol. 137, no. 5, pp. 835–48, 2009.