Understanding gene regulatory networks (GRNs) is crucial for elucidating cellular mechanisms and advancing therapeutic interventions. Original methods for GRN inference from bulk expression data often struggled with the high dimensionality and inherent noise in the data. Here we introduce RegDiffusion, a new class of Denoising Diffusion Probabilistic Models focusing on the regulatory effects among feature variables. RegDiffusion introduces Gaussian noise to the input data following a diffusion schedule and uses a neural network with a parameterized adjacency matrix to predict the added noise. We show that using this process, GRNs can be learned effectively with a surprisingly simple model architecture. In our benchmark experiments, RegDiffusion shows superior performance compared to several baseline methods in multiple datasets. We also demonstrate that RegDiffusion can infer biologically meaningful regulatory networks from real-world single-cell data sets with over 15,000 genes in under 5 minutes. This work not only introduces a fresh perspective on GRN inference but also highlights the promising capacity of diffusion-based models in the area of single-cell analysis. The RegDiffusion software package and experiment data are available at https://github.com/TuftsBCB/RegDiffusion.
Get full access to this article
View all access options for this article.
References
1.
ArrudaLB, SimD, ChikhlikarPR, et al.Dendritic cell-lysosomal-associated membrane protein (LAMP) and LAMP-1-HIV-1 gag chimeras have distinct cellular trafficking pathways and prime T and B cell responses to a diverse repertoire of epitopes. J Immunol, 2006; 177(4):2265–2275; doi: 10.4049/jimmunol.177.4.2265
2.
AytonS, FauxNG, BushAI, Alzheimer’s Disease Neuroimaging Initiative. Ferritin levels in the cerebrospinal fluid predict Alzheimer’s disease outcomes and are regulated by APOE. Nat Commun, 2015; 6(1):6760; doi: 10.1001/jamaneurol.2016.4406
3.
BalboulaAZ, SteinP, SchultzRM, et al.RBBP4 regulates histone deacetylation and bipolar spindle assembly during oocyte maturation in the mouse. Biol Reprod, 2015; 92(4):105; doi: 10.1095/biolreprod.115.128298
4.
BatorskyR, CeasrineAM, ShookLL, et al.Hofbauer cells and fetal brain microglia share transcriptional profiles and responses to maternal diet-induced obesity. bioRxiv, 2024; 43(6):114326; doi: 10.1101/2023.12.16.571680
5.
BiundoF, IshiwariK, Del PreteD, et al.Interaction of ApoE3 and ApoE4 isoforms with an ITM2b/BRI2 mutation linked to the Alzheimer disease-like Danish dementia: Effects on learning and memory. Neurobiol Learn Mem, 2015; 126:18–30; doi: 10.1016/j.nlm.2015.10.009
6.
ChanTE, StumpfMP, BabtieAC. Gene regulatory network inference from single-cell data using multivariate information measures. Cell Syst, 2017; 5(3):251–267.e3; doi: 10.1101/082099
7.
DavidsonE, LevinM. Gene regulatory networks. Proc Natl Acad Sci U S A, 2005; 102(14):4935–4935; doi: 10.1073/pnas.0502024102
8.
DavisJ, GoadrichM. The relationship between precision-recall and roc curves. In Proceedings of the 23rd International Conference on Machine Learning; 2006; pp. 233–240. 10.1145/1143844.1143874
9.
de Saint-VisB, VincentJ, VandenabeeleS, et al.A novel lysosome-associated membrane glycoprotein, DC-LAMP, induced upon DC maturation, is transiently expressed in MHC class ii compartment. Brain Behav Immun Integr, 1998; 9(3):325–336; doi: 10.1016/s1074-7613(00)80615-9
10.
Emmert-StreibF, DehmerM, Haibe-KainsB. Gene regulatory networks and their applications: Understanding biological and medical problems in terms of networks. Front Cell Dev Biol, 2014; 2:38; doi: 10.3389/fcell.2014.00038
11.
FaithJJ, HayeteB, ThadenJT, et al.Large-scale mapping and validation of Escherichia coli transcriptional regulation from a compendium of expression profiles. PLoS Biol, 2007; 5(1):e8; doi: 10.1371/journal.pbio.0050008
12.
Fernández-CalleR, KoningsSC, Frontiñán-RubioJ, et al.APOE in the bullseye of neurodegenerative diseases: Impact of the APOE genotype in Alzheimer’s disease pathology and brain diseases. Mol Neurodegener, 2022; 17(1):62; doi: 10.1186/s13024-022-00566-4
13.
FriedmanN, LinialM, NachmanI, et al.Using bayesian networks to analyze expression data. In Proceedings of the Fourth Annual International Conference on Computational Molecular Biology. 2000; pp. 127–135. 10.1145/332306.332355
14.
FuksF, BurgersWA, BrehmA, et al.DNA methyltransferase dnmt1 associates with histone deacetylase activity. Nat Genet, 2000; 24(1):88–91; doi: 10.1038/71750
15.
Garcia-AlonsoL, IbrahimMM, TureiD, et al.Benchmark and integration of resources for the estimation of human transcription factor activities. Genome Res, 2019; 29(8):1363–1375; doi: 10.1101/337915
16.
GhazanfarS, BisogniAJ, OrmerodJT, et al.Integrated single cell data analysis reveals cell specific networks and novel coactivation markers. BMC Syst Biol, 2016; 10(Suppl 5):127; doi: 10.1186/s12918-016-0370-4
17.
GuK, GreydanusS, MetzL, et al.Deep unsupervised learning using nonequilibrium thermodynamics. In International Conference on Machine Learning. PMLR; 2015; pp. 2256–2265. 10.1101/2019.12.30.891184
18.
HabenichtLKL, WangZ, ZhangX, et al.The C1q-ApoE complex: A new hallmark pathology of viral hepatitis and nonalcoholic fatty liver disease. Front Immunol, 2022; 13:970938; doi: 10.3389/fimmu.2022.970938
19.
HammondTR, DufortC, Dissing-OlesenL, et al.Single-cell RNA sequencing of microglia throughout the mouse lifespan and in the injured brain reveals complex cell-state changes. Brain Behav Immun Integr, 2019; 50(1):253–271.e6; doi: 10.1101/406140
20.
HanH, ChoJ, LeeS, et al.TRRUST v2: An expanded reference database of human and mouse transcriptional regulatory interactions. Nucleic Acids Res, 2018; 46(D1):D380–D386; doi: 10.1093/nar/gkx1013
21.
HeckerM, LambeckS, ToepferS, et al.Gene regulatory network inference: Data integration in dynamic models—a review. Biosystems, 2009; 96(1):86–103; doi: 10.1016/j.biosystems.2008.12.004
22.
HurJ, FrostGR, WuX, et al.The innate immunity protein IFITM3 modulates γ-secretase in Alzheimer’s disease. Npj Unconv Comput, 2020; 586(7831):735–740; doi: 10.1038/s41586-020-2681-2
23.
Huynh-ThuVA, GeurtsP. dyngenie3: Dynamical genie3 for the inference of gene networks from time series expression data. Sci Rep, 2018; 8(1):3384; doi: 10.1038/s41598-018-21715-0
24.
Huynh-ThuVA, IrrthumA, WehenkelL, et al.Inferring regulatory networks from expression data using tree-based methods. PLoS One, 2010; 5(9):e12776; doi: 10.1371/journal.pone.0012776
JanssenP, KliesmeteZ, ViethB, et al.The effect of background noise and its removal on the analysis of single-cell expression data. Genome Biol, 2023; 24(1):140; doi: 10.1101/2022.11.16.516780
27.
KangY, ThieffryD, CantiniL. Evaluating the reproducibility of single-cell gene regulatory network inference algorithms. Front Genet, 2021; 12:617282; doi: 10.1101/2020.11.10.375923
28.
KarlebachG, ShamirR. Modelling and analysis of gene regulatory networks. Nat Rev Mol Cell Biol, 2008; 9(10):770–780; doi: 10.1038/nrm2503
29.
KoldamovaR, LefterovI, FitzNF, et al.The role of APOE and trem2 in Alzheimer’s disease—current understanding and perspectives. IJMS, 2018; 20(1):81; doi: 10.1016/j.jalz.2018.06.2454
KrizhevskyA, SutskeverI, HintonGE. Dropout: A simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research, 2014; 15(1):1929–1958; doi: 10.1145/3065386
32.
LiH, ZhaoQ, LiuD, et al.Cathepsin B aggravates atherosclerosis in ApoE-deficient mice by modulating vascular smooth muscle cell pyroptosis through NF-κb/NLRP3 signaling pathway. PLoS One, 2024; 19(1):e0294514; doi: 10.1371/journal.pone.0294514
33.
LiuZ, WuC, MiaoH, et al.Regnetwork: An integrated database of transcriptional and post-transcriptional regulatory networks in human and mouse. Database (Oxford), 2015; 2015; doi: 10.1093/database/bav095
34.
LugerK, MäderAW, RichmondRK, et al.Crystal structure of the nucleosome core particle at 2.8 å resolution. Npj Unconv Comput, 1997; 389(6648):251–260; doi: 10.1038/38444
35.
MargolinAA, NemenmanI, BassoK, et al.ARACNE: An algorithm for the reconstruction of gene regulatory networks in a mammalian cellular context. BMC Bioinformatics, 2006; 7,(Suppl 1):S7–S15; doi: 10.1186/1471-2105-7-s1-s7
36.
MatsumotoH, KiryuH, FurusawaC, et al.SCODE: An efficient regulatory network inference algorithm from single-cell RNA-seq during differentiation. Bioinform Biomed Eng, 2017; 33(15):2314–2321; doi: 10.1101/088856
MeyerPE, KontosK, LafitteF, et al.Information-theoretic inference of large transcriptional regulatory networks. EURASIP J Bioinform Syst Biol, 2007; 2007(1):79879–79879; doi: 10.1155/2007/79879
39.
MoermanT, Aibar SantosS, Bravo González-BlasC, et al.GRNBoost2 and Arboreto: Efficient and scalable inference of gene regulatory networks. Bioinformatics, 2019; 35(12):2159–2161; doi: 10.1093/bioinformatics/bty916
40.
NguyenH, TranD, TranB, et al.A comprehensive survey of regulatory network inference methods using single cell RNA sequencing data. Brief Bioinform, 2021; 22(3):bbaa190; doi: 10.1093/bib/bbaa190
41.
OzgencilM, DulloviA, Christiane HigosRC, et al.The linker histone h1–brca1 axis is a crucial mediator of replication fork stability. Life Sci Alliance, 2023; 6(9); doi: 10.26508/lsa.202301933
RauA, JaffrézicF, FoulleyJ, et al.An empirical bayesian method for estimating biological networks from temporal microarray data. Stat Appl Genet Mol Biol, 2010; 9(1):Article 9; doi: 10.2202/1544-6115.1513
47.
RonnebergerO, FischerP, BroxT. U-Net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. Springer; 2015; pp. 234–241; doi: 10.1007/978-3-319-24574-4_28
48.
RussoA, CordelliE, SalvittiT, et al.Rad54/Rad54B deficiency is associated to increased chromosome breakage in mouse spermatocytes. J Carcinog Mutagen, 2018; 33(4):323–332; doi: 10.1093/mutage/gey027
49.
KimS. ppcor: An r package for a fast calculation to semi-partial correlation coefficients. Commun Stat Appl Methods, 2015; 22(6):665–674; doi: 10.5351/csam.2015.22.6.665
50.
LèbreS. Inferring dynamic genetic networks with low order independencies. Stat Appl Genet Mol Biol, 2009; 8(1):Article 9; doi: 10.2202/1544-6115.1294
Sala FrigerioC, WolfsL, FattorelliN, et al.The major risk factors for Alzheimer’s disease: Age, sex, and genes modulate the microglia response to a β plaques. Cell Rep, 2019; 27(4):1293–1306.e6; doi: 10.1016/j.celrep.2019.03.099
53.
SamokhinAO, LythgoPA, GauthierJY, et al.Pharmacological inhibition of cathepsin S decreases atherosclerotic lesions in Apoe-/-mice. J Cardiovasc Pharmacol, 2010; 56(1):98–105; doi: 10.1097/fjc.0b013e3181e23e10
54.
Sanchez-CastilloM, BlancoD, Tienda-LunaIM, et al.A bayesian framework for the inference of gene regulatory networks from time and pseudo-time series data. Bioinformatics, 2018; 34(6):964–970; doi: 10.1093/bioinformatics/btx605
55.
SchaffterT, MarbachD, FloreanoD. GeneNetWeaver: In silico benchmark generation and performance profiling of network inference methods. Bioinformatics, 2011; 27(16):2263–2270; doi: 10.1093/bioinformatics/btr373
56.
ShuH, ZhouJ, LianQ, et al.Modeling gene regulatory networks using neural network architectures. Nat Comput Sci, 2021; 1(7):491–501; doi: 10.1038/s43588-021-00099-8
57.
SierksmaA, LuA, SaltaE, et al.but not to TAU pathology. EMBO Mol Med, 2020; 12(3):e10606; doi: 10.1101/491902
58.
SlonimDK. From patterns to pathways: Gene expression data analysis comes of age. Nat Genet, 2002; 32(4):502–508; doi: 10.1038/ng1033
59.
SongY, DurkanC, MurrayI, et al.Maximum likelihood training of score-based diffusion models. Adv Neural Inf Process Syst, 2021; 34:1415–1428.
60.
SzklarczykD, GableAL, LyonD, et al.String v11: Protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res, 2019; 47(D1):D607–D613; doi: 10.1093/nar/gky1131
61.
TangW, LiuR, WenH, et al.A general single-cell analysis framework via conditional diffusion generative models. bioRxiv, 2023; doi: 10.1101/2023.10.13.562243
62.
TiribuziR, OrlacchioA, CrispoltoniL, et al.Lysosomal β-galactosidase and β-hexosaminidase activities correlate with clinical stages of dementia associated with Alzheimer’s disease and type 2 diabetes mellitus. J Alzheimers Dis, 2011; 24(4):785–797; doi: 10.3233/jad-2011-100525
63.
van DamS, VõsaU, van der GraafA, et al.Gene co-expression analysis for functional classification and gene–disease predictions. Brief Bioinform, 2018; 19(4):575–592; doi: 10.1093/bib/bbw139
64.
YadatiT, HoubenT, BitorinaA, et al.The ins and outs of cathepsins: Physiological function and role in disease management. Cells Dev, 2020; 9(7):1679; doi: 10.3390/cells9071679
65.
ZalocuskyKA, NajmR, TaubesAL, et al.Neuronal ApoE upregulates MHC-I expression to drive selective neurodegeneration in Alzheimer’s disease. Nat Neurosci, 2021; 24(6):786–798; doi: 10.1038/s41593-021-00851-3
66.
ZhouW, ScottS, SheltonS, et al.Cathepsin D-mediated proteolysis of apolipoprotein E: Possible role in Alzheimer’s disease. NPP Digit Psychiatry Neurosci, 2006; 143(3):689–701; doi: 10.1016/j.neuroscience.2006.08.019
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
