Study of N6-methyladenosine using tensor decomposition-based unsupervised feature extraction

Transcript

1. isaic2021 1 Study of N6-methyladenosine using tensor decomposition-based unsupervised feature

   extraction Y-h. Taguchi, Department of Physics, Chuo University, Tokyo Japan.

2. isaic2021 2 The content of this talk is based upon

   the following two papers Taguchi, Y.-h.; Dharshini, S.A.P.; Gromiha, M.M. Identification of Transcription Factors, Biological Pathways, and Diseases as Mediated by N6-methyladenosine Using Tensor Decomposition-Based Unsupervised Feature Extraction. Appl. Sci. 2021, 11, 213. https://doi.org/10.3390/app11010213 Roy, S.S., Taguchi, YH. Identification of genes associated with altered gene expression and m6A profiles during hypoxia using tensor decomposition based unsupervised feature extraction. Sci Rep 11, 8909 (2021). https://doi.org/10.1038/s41598-021-87779-7

3. isaic2021 3 What is m6A? = Methylation of Adenosine Adenosine

   Methylation

4. isaic2021 4 m6A is the most ubiquitous methylation of RNA

   in eukaryotes. It is also known to be related to various biological functions including • Alternative splicing • Transfer from nuclei to cytoplasm • Translation • Stability • Transcription That is, m6A affects almost all aspects of RNA.

5. isaic2021 5 Paramasivam, A., Vijayashree Priyadharsini, J. & Raghunandhakumar, S.

   N6-adenosine methylation (m6A): a promising new molecular target in hypertension and cardiovascular diseases. Hypertens Res 43, 153–154 (2020). How does m6A occur and vanish?

6. isaic2021 6 At the moment, we do not know about

   the function of m6A well. We discuss the following two topics of m6A in this talk. 1. How does m6A affect other biological factors, e.g., transcription factors, biological pathways, and diseases ? 2. Specifically, how does m6A affect hypoxia ? Method: Tensor decomposition

7. isaic2021 7 What is a tensor? Scholar x: a number

   Vector x i : a set of scholars in line Matrix x ij : a set of scholars aligned in a table (i.e. rows and columns) Tensor x ijk : a set of scholars aligned in an array more then two rows x ijk i j k 1 (1,2,3,4,...) (1 2 3 4 5 6 7 8 9 )

8. isaic2021 8 Tensor is suitable to store genomics data: Gene

   expression :x ijk ∈ ℝN⨉M⨉K N genes ⨉ M persons ⨉ K tissues x ijk i:genes j:persons k:tissues

9. isaic2021 9 What is tensor decomposition(TD)? Expand tensor as a

   series of product of vectors, x ijk i:genes j:persons k:tissues G k j i l 1 l 2 l 3 = u l 1 i u l 2 j u l 3 k u l 1 i u l 2 j u l 3 k x ijk ≃∑ l 1 =1 L 1 ∑ l 2 =2 L 2 ∑ l 3 =1 L 3 G (l 1 l 2 l 3 )u l 1 i u l 2 j u l 3 k

10. isaic2021 10 Advantages of tensor decomposition(TD): We can know “Dependence

    of x ijk upon i” → u l1i “Dependence of x ijk upon j” → u l2j “Dependence of x ijk upon k” → u l3k ← Healthy control vs patient ← tissue specificity Gene selection ↑ We can answer the question : Which genes are expressed between healthy controls and patients in tissue specific manner?

11. isaic2021 11 Interpretation….. j:samples Healthy control Patients ul2j For some

    specific l2 For some specific l3 k:tissues Tissue specific expression ul3k

12. isaic2021 12 i:genes ul1i tDEG: tissue specific Differentially Expressed Genes

    Healthy controls < Patients tDEG: tDEG: Healthy controls > Patients For some specific l1 with max |G(l1l2l3)| If G(l1l2l3)>0 Fixed

13. isaic2021 13 1. How does m6A affect other biological factors,

    e.g., 1. How does m6A affect other biological factors, e.g., t transcription factors, biological pathways, and diseases ? ranscription factors, biological pathways, and diseases ?

14. isaic2021 14 Four features, two histone modifications (H3K27ac, H3K4me3), m6A

    and gene expression, were measured with and without METTL3 KO. They are attributed to genes as well. Purpose: Purpose: Which genes are associated with histone modifications, m6A and gene expression, altered by METTL3 KO?

15. isaic2021 15 Regardless to features, altered by METTL3 KO, common

    between replicates Human Human

16. isaic2021 16 Regardless to features, altered by METTL3 KO, common

    between replicates Mouse Mouse

17. isaic2021 17 Genes selected between human and mouse are significantly

    overlapped. Odds ratio = 4.0, P-values = 1.91 ⨉ 10-17

18. isaic2021 18 TFs that significantly regulate selected genes are significantly

    overlapped between human and mouse. The fact that majority of TFs are commonly selected is remarkable, since only 66 genes only 66 genes are common between 664 human genes and 613 mouse genes.

19. isaic2021 19 Diseases with up/down regulated selected genes are also

    common between human and mouse

20. isaic2021 20 Ten and four commonly selected diseases Ten and

    four commonly selected diseases Disease Perturbations from GEO down Disease Perturbations from GEO down "Crohn’s disease DOID-8778 human GSE6731 sample 757"; "Ulcerative Colitis C0009324 human GSE6731 sample 249"; "Ulcerative colitis DOID-8577 human GSE6731 sample 759"; "Ulcerative colitis DOID-8577 human GSE6731 sample 760"; "Hepatitis C DOID-1883 human GSE20948 sample 599"; "Diabetic Nephropathy C0011881 human GSE1009 sample 223"; "Cardiac Hypertrophy C1383860 rat GSE1055 sample 354"; "Prostate cancer DOID-10283 human GSE3868 sample 639"; "Hepatitis C DOID-1883 human GSE20948 sample 597"; "Cancer of prostate C0376358 human GSE3868 sample 135" Disease Perturbations from GEO up Disease Perturbations from GEO up "Pancreatitis DOID-4989 mouse GSE3644 sample 513 Asthma"; "Allergic C0155877 human GSE3004 sample 360"; "Acute pancreatitis C0001339 mouse GSE3644 sample 376"; "Chronic lymphocytic leukemia DOID-1040 human GSE6691 sample 786"

21. isaic2021 21 Conclusion in part 1 Conclusion in part 1

    Using TD, we could identify common genes associated with histon modification, m6A, gene expression altered with METTL3 KO, and common TFs that regulated these genes and common diseases associated with altered expression of these genes between human and mouse.

22. isaic2021 22 2. Specifically, how does m6A affect 2. Specifically,

    how does m6A affect hypoxia ? hypoxia ?

23. isaic2021 23 GSE141941 Gene expression and m6A profiles at four

    time points in response to hypoxia.

24. isaic2021 24 Gene expression and m6A profiles were first treated

    with principal component analysis (PCA) and TD separately, then are integrated with kernel TD.

25. isaic2021 25 PC loading attributed to time points when PCA

    was applied to gene expression gene expression Singular value vectors attributed to time points when TD was applied to m6A profiles m6A profiles 2nd ones are correlated with times and are common between gene expression and m6A profiles

26. isaic2021 26 Integrated analysis of m6A profiles and gene expression

    by kernel TD also gives 2nd vectors attributed to times and correlated between gene expression and m6A profiles. Gene expression M6A profiles Scatter plot of 2nd ones

27. isaic2021 27 “KEGG 2019 Human” category of Enrichr for 53

    genes (based on gene expression) and 200 genes (based on m6A) selected by applying KTD-based unsupervised FE to integration of gene expression and m6A profile. Fifteen terms with adjusted P-values less than 0.05 are listed. Hypoxia-inducible factor 1 Hypoxia-inducible factor 1

28. isaic2021 28 Conclusion in part 2 Conclusion in part 2

    We have successfully integrated gene expression and m6A profiles at four time points in response to hypoxia. Genes associated with time development are related to HIF-1 signaling pathway. m6A likely regulated hipoxia through gene expression regulated by m6A.