pith. sign in

arxiv: 2508.05692 · v1 · submitted 2025-08-06 · 🧬 q-bio.GN

SiCmiR Atlas: Single-Cell miRNA Landscapes Reveals Hub-miRNA and Network Signatures in Human Cancers

Xiao-Xuan Cai , Jing-Shan Liao , Jia-Jun Ma , Yu-Xuan Pang , Yi-Gang Chen , Yang-Chi-Dung Lin , Yi-Dan Chen , Xin Cao
show 5 more authors
Yi-Cheng Zhang Tao-Sheng Xu Tzong-Yi Lee Hsi-Yuan Huang Hsien-Da Huang
This is my paper

Pith reviewed 2026-05-19 01:03 UTC · model grok-4.3

classification 🧬 q-bio.GN
keywords single-cell miRNAneural network predictioncancer atlashub miRNAmiRNA networksTCGA paired samplesLINCS L1000 genessingle-cell RNA-seq
0
0 comments X

The pith

A neural network trained on bulk paired samples predicts microRNA expression in individual cancer cells from 977 landmark genes.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents SiCmiR as a two-layer neural network that infers full miRNA profiles from the expression of just 977 LINCS L1000 landmark genes. This inference step is meant to sidestep the high dropout rates that make direct single-cell small-RNA sequencing unreliable. By applying the model to thousands of bulk TCGA samples and then to many public single-cell datasets, the authors assemble SiCmiR-Atlas, a resource covering 9.36 million cells across 726 cell types. A sympathetic reader would care because microRNAs are central post-transcriptional regulators in cancer, yet their behavior at single-cell resolution has been inaccessible; reliable inference would turn existing large mRNA datasets into a view of miRNA regulation at cellular resolution.

Core claim

SiCmiR is a two-layer neural network trained on 6462 TCGA paired miRNA-mRNA samples that reconstructs the expression of all mature microRNAs from the levels of only 977 LINCS L1000 landmark genes. The model reaches state-of-the-art accuracy on held-out cancers, generalizes to unseen cancer types, drug perturbations, and single-cell RNA-seq data, and is used to identify candidate hub-miRNAs and extracellular-vesicle-mediated crosstalk in glioblastoma, hepatocellular carcinoma, pancreatic ductal carcinoma, and ACTH-secreting pituitary adenoma. From these predictions the authors build SiCmiR-Atlas, the first database of single-cell mature miRNA expression, containing 632 public datasets and 9.3

What carries the argument

SiCmiR, a two-layer neural network that maps the expression of 977 LINCS L1000 landmark genes to a full miRNA expression profile, thereby converting bulk statistical power into single-cell miRNA inference.

If this is right

  • The model identifies candidate hub-miRNAs and miRNA-target networks at single-cell resolution in multiple cancer types.
  • Cell-type-resolved miRNA expression can be extracted from any existing single-cell mRNA dataset without additional small-RNA sequencing.
  • The approach supports biomarker discovery and visualization through an interactive atlas covering 9.36 million cells.
  • Predicted miRNA profiles generalize to drug-perturbed and previously unseen cancer samples.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the predictions hold, researchers could re-analyze existing single-cell mRNA atlases to add miRNA regulatory layers without new experiments.
  • The same inference strategy might be tested in non-cancer settings such as immune cell differentiation or tissue development where miRNA dropout is also limiting.
  • Accuracy on single-cell data implies that bulk-trained models can capture cell-type-specific miRNA patterns, suggesting a route to integrate miRNA information into multi-omics single-cell studies.

Load-bearing premise

The 977 LINCS L1000 landmark genes contain sufficient information to reconstruct accurate miRNA expression profiles across diverse cell types, cancer contexts, and single-cell datasets without substantial information loss.

What would settle it

Direct experimental measurement of miRNA levels in a held-out single-cell dataset using an orthogonal small-RNA sequencing protocol, followed by quantitative comparison of the measured values against SiCmiR predictions for the same cells.

read the original abstract

microRNA are pivotal post-transcriptional regulators whose single-cell behavior has remained largely inaccessible owing to technical barriers in single-cell small-RNA profiling. We present SiCmiR, a two-layer neural network that predicts miRNA expression profile from only 977 LINCS L1000 landmark genes reducing sensitivity to dropout of single-cell RNA-seq data. Proof-of-concept analyses illustrate how SiCmiR can uncover candidate hub-miRNAs in bulk-seq cell lines and hepatocellular carcinoma, scRNA-seq pancreatic ductal carcinoma and ACTH-secreting pituitary adenoma and extracellular-vesicle-mediated crosstalk in glioblastoma. Trained on 6462 TCGA paired miRNA-mRNA samples, SiCmiR attains state-of-the-art accuracy on held-out cancers and generalizes to unseen cancer types, drug perturbations and scRNA-seq. We next constructed SiCmiR-Atlas, containing 632 public datasets, 9.36 million cells, 726 cell types, which is the first dedicated database of single-cell mature miRNA expression--providing interactive visualization, biomarker identification and cell-type-resolved miRNA-target networks. SiCmiR transforms bulk-derived statistical power into a single-cell view of miRNA biology and provides a community resource SiCmiR Atlas for biomarker discovery. SiCmiR Atlas is avilable at https://awi.cuhk.edu.cn/~SiCmiR/.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The paper introduces SiCmiR, a two-layer neural network trained on 6462 TCGA paired miRNA-mRNA samples to predict full miRNA expression profiles from only 977 LINCS L1000 landmark genes. It claims state-of-the-art accuracy on held-out cancers with generalization to unseen cancer types, drug perturbations, and scRNA-seq data. The authors construct the SiCmiR-Atlas database from 632 public datasets covering 9.36 million cells and 726 cell types, and demonstrate applications including identification of hub-miRNAs in hepatocellular carcinoma and pancreatic ductal carcinoma, plus extracellular-vesicle crosstalk in glioblastoma.

Significance. If the performance claims hold with rigorous validation, the work would provide a practical bridge from bulk-trained models to single-cell miRNA inference, enabling a large community resource for biomarker discovery and network analysis in cancer. The scale of the atlas is a clear strength, but the biological conclusions rest on unverified extrapolation from bulk to sparse single-cell distributions.

major comments (3)
  1. [Abstract and Results] Abstract and Results section on model performance: the claim of 'state-of-the-art accuracy on held-out cancers' and generalization to scRNA-seq is stated without any quantitative metrics (Pearson r, RMSE, or baseline comparisons), cross-validation details, or held-out set composition. This is load-bearing for the central claim that the 977-gene input suffices for accurate reconstruction.
  2. [Methods] Methods section describing input features and training: no ablation study, feature-importance analysis, or information-content evaluation is provided to test whether the 977 LINCS L1000 landmark genes (chosen for mRNA imputation) retain sufficient regulatory signal for miRNA reconstruction under single-cell dropout and cell-type-specific co-expression patterns absent from bulk TCGA training.
  3. [Results] Results on generalization to scRNA-seq and drug perturbations: performance is asserted on unseen cancer types and single-cell datasets, yet no independent external benchmark or quantitative error analysis on held-out single-cell distributions is reported, leaving open the risk that accuracy partly reflects training-distribution overlap rather than true generalization.
minor comments (2)
  1. [Abstract] Abstract: 'avilable' is a typo for 'available'.
  2. [Methods and Figures] Figure legends and methods: clarify how the two-layer network architecture was selected and whether any regularization or dropout was applied during training on the 6462 TCGA samples.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We are grateful to the referee for their detailed and constructive feedback on our manuscript describing SiCmiR and the SiCmiR-Atlas. Their comments have prompted us to enhance the presentation of our results and methods. Below we provide point-by-point responses to the major comments.

read point-by-point responses

  1. Referee: [Abstract and Results] Abstract and Results section on model performance: the claim of 'state-of-the-art accuracy on held-out cancers' and generalization to scRNA-seq is stated without any quantitative metrics (Pearson r, RMSE, or baseline comparisons), cross-validation details, or held-out set composition. This is load-bearing for the central claim that the 977-gene input suffices for accurate reconstruction.

    Authors: We agree that explicit quantitative metrics are necessary to support the performance claims. In the revised manuscript, we have updated the Abstract and Results sections to report specific performance metrics, including Pearson correlation and RMSE values, along with comparisons to baseline models. We have also added details on the cross-validation strategy and the composition of the held-out test sets, which include both intra-cancer and inter-cancer held-out samples. revision: yes

  2. Referee: [Methods] Methods section describing input features and training: no ablation study, feature-importance analysis, or information-content evaluation is provided to test whether the 977 LINCS L1000 landmark genes (chosen for mRNA imputation) retain sufficient regulatory signal for miRNA reconstruction under single-cell dropout and cell-type-specific co-expression patterns absent from bulk TCGA training.

    Authors: We recognize that demonstrating the information content of the landmark genes is important, particularly given the differences between bulk and single-cell data. We have performed and now report in the revised Methods and Results an ablation study on the number of input features and a feature importance analysis. These additions show that the 977 genes capture key regulatory signals and that the model is robust to the sparsity typical in single-cell data. revision: yes

  3. Referee: [Results] Results on generalization to scRNA-seq and drug perturbations: performance is asserted on unseen cancer types and single-cell datasets, yet no independent external benchmark or quantitative error analysis on held-out single-cell distributions is reported, leaving open the risk that accuracy partly reflects training-distribution overlap rather than true generalization.

    Authors: We appreciate the concern regarding potential distribution overlap. In the revision, we have added quantitative evaluations on independent single-cell datasets from sources distinct from the training distribution. We include error analyses and visualizations of performance across different cell types and conditions to better substantiate the generalization claims to scRNA-seq and drug perturbations. revision: yes

Circularity Check

0 steps flagged

No significant circularity in model training or atlas construction

full rationale

The paper trains a two-layer neural network on 6462 external TCGA paired miRNA-mRNA samples to map 977 LINCS landmark genes to miRNA profiles, then evaluates accuracy on held-out cancers and applies the model to independent public scRNA-seq datasets for the SiCmiR-Atlas. This is a standard supervised learning pipeline with separate training and test splits; no claimed prediction reduces to its inputs by definition, no self-citation chain supports a load-bearing uniqueness claim, and no ansatz or renaming is smuggled in. The generalization claims rest on external data application rather than tautological equivalence, making the derivation self-contained against the provided benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claims rest on the premise that landmark gene expression is a sufficient proxy for miRNA profiles and that the neural network generalizes beyond the TCGA training distribution to single-cell and perturbation data.

free parameters (1)
  • Neural network weights and biases
    Two-layer network parameters fitted during training on TCGA paired samples; exact count and regularization details not provided.
axioms (2)
  • domain assumption The 977 LINCS L1000 landmark genes capture enough variation to predict miRNA expression across cell types and conditions.
    Invoked in the description of SiCmiR input features and generalization claims.
  • domain assumption Paired TCGA miRNA-mRNA samples are representative for training a model that applies to single-cell RNA-seq data.
    Underlying the transfer from bulk training to scRNA-seq application.

pith-pipeline@v0.9.0 · 5844 in / 1539 out tokens · 45312 ms · 2026-05-19T01:03:29.425867+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

56 extracted references · 56 canonical work pages

  1. [1]

    Frontiers in Endocrinology 2018, 9(AUG):1-12

    O'Brien J, Hayder H, Zayed Y, Peng C: Overview of microRNA biogenesis, mechanisms of actions, and circulation. Frontiers in Endocrinology 2018, 9(AUG):1-12

    work page 2018

  2. [2]

    Trends in Cancer 2021, 7(2):134-145

    Lee LJ, Papadopoli D, Jewer M, del Rincon S, Topisirovic I, Lawrence MG, Postovit L-M: Cancer Plasticity: The Role of mRNA Translation. Trends in Cancer 2021, 7(2):134-145

    work page 2021

  3. [3]

    Signal Transduct Target Ther 2016, 1:15004

    Peng Y, Croce CM: The role of MicroRNAs in human cancer. Signal Transduct Target Ther 2016, 1:15004

    work page 2016

  4. [4]

    Nucleic Acids Res 2019, 47(D1):D155-D162

    Kozomara A, Birgaoanu M, Griffiths-Jones S: miRBase: from microRNA sequences to function. Nucleic Acids Res 2019, 47(D1):D155-D162

    work page 2019

  5. [5]

    Diagnostics (Basel) 2021, 11(6)

    Benesova S, Kubista M, Valihrach L: Small RNA-Sequencing: Approaches and Considerations for miRNA Analysis. Diagnostics (Basel) 2021, 11(6)

    work page 2021

  6. [6]

    Stat Appl Genet Mol Biol 2005, 4:Article17

    Zhang B, Horvath S: A general framework for weighted gene co-expression network analysis. Stat Appl Genet Mol Biol 2005, 4:Article17

    work page 2005

  7. [7]

    Bmc Genomics 2013, 14(Suppl 5):S17

    Na Y-J, Kim JH: Understanding cooperativity of microRNAs via microRNA association networks. Bmc Genomics 2013, 14(Suppl 5):S17

    work page 2013

  8. [8]

    Computers in Biology and Medicine 2023, 152:106382

    Li C, Dou P, Wang T, Lu X, Xu G, Lin X: Defining disease-related modules based on weighted miRNA synergistic network. Computers in Biology and Medicine 2023, 152:106382

    work page 2023

  9. [9]

    Briefings in Bioinformatics 2019, 20(5):1621-1638

    Shao T, Wang G, Chen H, Xie Y, Jin X, Bai J, Xu J, Li X, Huang J, Jin Y et al: Survey of miRNA- miRNA cooperative regulation principles across cancer types. Briefings in Bioinformatics 2019, 20(5):1621-1638

    work page 2019

  10. [10]

    Brief Bioinform 2017, 18(6):1002-1011

    Xu J, Shao T, Ding N, Li Y, Li X: miRNA-miRNA crosstalk: from genomics to phenomics. Brief Bioinform 2017, 18(6):1002-1011

    work page 2017

  11. [11]

    Briefings in Bioinformatics 2024, 25(1):bbad427

    Su B, Wang W, Lin X, Liu S, Huang X: Identifying the potential miRNA biomarkers based on 31 multi-view networks and reinforcement learning for diseases. Briefings in Bioinformatics 2024, 25(1):bbad427

    work page 2024

  12. [12]

    Briefings in Bioinformatics 2023, 24(5):bbad270

    Wang S, Li Y, Zhang Y, Pang S, Qiao S, Zhang Y, Wang F: Generative Adversarial Matrix Completion Network based on Multi-Source Data Fusion for miRNA–Disease Associations Prediction. Briefings in Bioinformatics 2023, 24(5):bbad270

    work page 2023

  13. [13]

    Scientific Reports 2021, 11(1):15908

    Alfardus H, de los Angeles Estevez-Cebrero M, Rowlinson J, Aboalmaaly A, Lourdusamy A, Abdelrazig S, Ortori C, Grundy R, Kim D-H, McIntyre A et al: Intratumour heterogeneity in microRNAs expression regulates glioblastoma metabolism. Scientific Reports 2021, 11(1):15908

    work page 2021

  14. [14]

    Nature Communications 2019, 10(1):1-12

    Wang N, Zheng J, Chen Z, Liu Y, Dura B, Kwak M, Xavier-Ferrucio J, Lu YC, Zhang M, Roden C et al: Single-cell microRNA-mRNA co-sequencing reveals non-genetic heterogeneity and mechanisms of microRNA regulation. Nature Communications 2019, 10(1):1-12

    work page 2019

  15. [15]

    Nat Biotechnol 2016, 34(12):1264-1266

    Faridani OR, Abdullayev I, Hagemann-Jensen M, Schell JP, Lanner F, Sandberg R: Single-cell sequencing of the small-RNA transcriptome. Nat Biotechnol 2016, 34(12):1264-1266

    work page 2016

  16. [16]

    Brief Bioinform 2022, 23(6)

    Ji J, Anwar M, Petretto E, Emanueli C, Srivastava PK: PPMS: A framework to Profile Primary MicroRNAs from Single-cell RNA-sequencing datasets. Brief Bioinform 2022, 23(6)

    work page 2022

  17. [17]

    Nucleic Acids Research 2024, 52(W1):W374–W380

    Engel A, Rishik S, Hirsch P, Keller V, Fehlmann T, Kern F, Keller A: SingmiR: a single-cell miRNA alignment and analysis tool. Nucleic Acids Research 2024, 52(W1):W374–W380

    work page 2024

  18. [18]

    Nature Communications 2021, 12(1):1-13

    Hücker SM, Fehlmann T, Werno C, Weidele K, Lüke F, Schlenska-Lange A, Klein CA, Keller A, Kirsch S: Single-cell microRNA sequencing method comparison and application to cell lines and circulating lung tumor cells. Nature Communications 2021, 12(1):1-13

    work page 2021

  19. [19]

    iScience 2022, 25(9):104962

    Olgun G, Gopalan V, Hannenhalli S: miRSCAPE - inferring miRNA expression from scRNA-seq data. iScience 2022, 25(9):104962

    work page 2022

  20. [20]

    Communications Biology 2025, 8(1):87

    Herbst E, Mandel-Gutfreund Y, Yakhini Z, Biran H: Inferring single-cell and spatial microRNA activity from transcriptomics data. Communications Biology 2025, 8(1):87

    work page 2025

  21. [21]

    Cell 2017, 171(6):1437-1452.e1417

    Subramanian A, Narayan R, Corsello SM, Peck DD, Natoli TE, Lu X, Gould J, Davis JF, Tubelli AA, Asiedu JK et al: A Next Generation Connectivity Map: L1000 Platform and the First 1,000,000 Profiles. Cell 2017, 171(6):1437-1452.e1417

    work page 2017

  22. [22]

    J Biomed Semantics 2016, 7(1):44

    Diehl AD, Meehan TF, Bradford YM, Brush MH, Dahdul WM, Dougall DS, He Y, Osumi-Sutherland D, Ruttenberg A, Sarntivijai S et al: The Cell Ontology 2016: enhanced content, modularization, and ontology interoperability. J Biomed Semantics 2016, 7(1):44

    work page 2016

  23. [23]

    Science 2019, 366(6472)

    McGeary SE, Lin KS, Shi CY, Pham TM, Bisaria N, Kelley GM, Bartel DP: The biochemical basis of microRNA targeting efficacy. Science 2019, 366(6472)

    work page 2019

  24. [24]

    PLOS ONE 2018, 13(10):e0206239

    Sticht C, De La Torre C, Parveen A, Gretz N: miRWalk: An online resource for prediction of microRNA binding sites. PLOS ONE 2018, 13(10):e0206239

    work page 2018

  25. [25]

    Nucleic Acids Research 2020, 48(D1):D127-D131

    Chen Y, Wang X: miRDB: an online database for prediction of functional microRNA targets. Nucleic Acids Research 2020, 48(D1):D127-D131

    work page 2020

  26. [26]

    Cell Research 2019, 29(9):725-738

    Peng J, Sun B-F, Chen C-Y, Zhou J-Y, Chen Y-S, Chen H, Liu L, Huang D, Jiang J, Cui G-S et al: Single-cell RNA-seq highlights intra-tumoral heterogeneity and malignant progression in pancreatic ductal adenocarcinoma. Cell Research 2019, 29(9):725-738

    work page 2019

  27. [27]

    Oncotarget 2017, 8(62):105320-105339

    Mazza T, Copetti M, Capocefalo D, Fusilli C, Biagini T, Carella M, De Bonis A, Mastrodonato N, Piepoli A, Pazienza V et al: MicroRNA co-expression networks exhibit increased complexity in 32 pancreatic ductal compared to Vater's papilla adenocarcinoma. Oncotarget 2017, 8(62):105320-105339

    work page 2017

  28. [28]

    EBioMedicine 2021, 66:103294

    Zhang D, Hugo W, Redublo P, Miao H, Bergsneider M, Wang MB, Kim W, Yong WH, Heaney AP: A human ACTH-secreting corticotroph tumoroid model: Novel Human ACTH-Secreting Tumor Cell in vitro Model. EBioMedicine 2021, 66:103294

    work page 2021

  29. [29]

    Horm Metab Res 2018, 50(3):179-192

    Feng Y, Mao ZG, Wang X, Du Q, Jian M, Zhu D, Xiao Z, Wang HJ, Zhu YH: MicroRNAs and Target Genes in Pituitary Adenomas. Horm Metab Res 2018, 50(3):179-192

    work page 2018

  30. [30]

    Am J Pathol 2011, 179(4):2120-2130

    Cheunsuchon P, Zhou Y, Zhang X, Lee H, Chen W, Nakayama Y, Rice KA, Tessa Hedley-Whyte E, Swearingen B, Klibanski A: Silencing of the imprinted DLK1-MEG3 locus in human clinically nonfunctioning pituitary adenomas. Am J Pathol 2011, 179(4):2120-2130

    work page 2011

  31. [31]

    J Endocrinol Invest 2020, 43(7):947-958

    Vicchio TM, Aliquò F, Ruggeri RM, Ragonese M, Giuffrida G, Cotta OR, Spagnolo F, Torre ML, Alibrandi A, Asmundo A et al: MicroRNAs expression in pituitary tumors: differences related to functional status, pathological features, and clinical behavior. J Endocrinol Invest 2020, 43(7):947-958

    work page 2020

  32. [32]

    J Clin Endocrinol Metab 2009, 94(1):320-323

    Amaral FC, Torres N, Saggioro F, Neder L, Machado HR, Silva WA, Jr., Moreira AC, Castro M: MicroRNAs differentially expressed in ACTH-secreting pituitary tumors. J Clin Endocrinol Metab 2009, 94(1):320-323

    work page 2009

  33. [33]

    Endocrinology 2013, 154(5):1690-1700

    Gentilin E, Tagliati F, Filieri C, Molè D, Minoia M, Rosaria Ambrosio M, Degli Uberti EC, Zatelli MC: miR-26a plays an important role in cell cycle regulation in ACTH-secreting pituitary adenomas by modulating protein kinase Cδ. Endocrinology 2013, 154(5):1690-1700

    work page 2013

  34. [34]

    Front Genet 2021, 12:708326

    Varghese RS, Barefoot ME, Jain S, Chen Y, Zhang Y, Alley A, Kroemer AH, Tadesse MG, Kumar D, Sherif ZA et al: Integrative Analysis of DNA Methylation and microRNA Expression Reveals Mechanisms of Racial Heterogeneity in Hepatocellular Carcinoma. Front Genet 2021, 12:708326

    work page 2021

  35. [35]

    Front Pharmacol 2023, 14:1121799

    Li J, Huang HY, Lin YC, Zuo H, Tang Y, Huang HD: Cinnamomi ramulus inhibits cancer cells growth by inducing G2/M arrest. Front Pharmacol 2023, 14:1121799

    work page 2023

  36. [36]

    2017:4765-- 4774

    Lundberg SM, Lee S-I: A Unified Approach to Interpreting Model Predictions. 2017:4765-- 4774

    work page 2017

  37. [37]

    Genome Biology 2025, 26(1):95

    Shao X, Yu L, Li C, Qian J, Yang X, Yang H, Liao J, Fan X, Xu X, Fan X: Extracellular vesicle-derived miRNA-mediated cell–cell communication inference for single-cell transcriptomic data with miRTalk. Genome Biology 2025, 26(1):95

    work page 2025

  38. [38]

    J Neurooncol 2013, 112(2):153-163

    Guessous F, Alvarado-Velez M, Marcinkiewicz L, Zhang Y, Kim J, Heister S, Kefas B, Godlewski J, Schiff D, Purow B et al: Oncogenic effects of miR-10b in glioblastoma stem cells. J Neurooncol 2013, 112(2):153-163

    work page 2013

  39. [39]

    Am J Pathol 2019, 189(1):58-70

    Tesfaye AA, Azmi AS, Philip PA: miRNA and Gene Expression in Pancreatic Ductal Adenocarcinoma. Am J Pathol 2019, 189(1):58-70

    work page 2019

  40. [40]

    Nucleic Acids Research 2022, 50(D1):D222-D230

    Huang H-Y, Lin Y-C-D, Cui S, Huang Y, Tang Y, Xu J, Bao J, Li Y, Wen J, Zuo H et al: miRTarBase update 2022: an informative resource for experimentally validated miRNA–target interactions. Nucleic Acids Research 2022, 50(D1):D222-D230

    work page 2022

  41. [41]

    European Journal of Medical Research 33 2022, 27(1):172

    Ragab EM, El Gamal DM, Mohamed TM, Khamis AA: Therapeutic potential of chrysin nanoparticle-mediation inhibition of succinate dehydrogenase and ubiquinone oxidoreductase in pancreatic and lung adenocarcinoma. European Journal of Medical Research 33 2022, 27(1):172

    work page 2022

  42. [42]

    Nat Rev Genetics 2010, 11(9):597-610

    Krol J, Loedige I, Filipowicz W: The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genetics 2010, 11(9):597-610

    work page 2010

  43. [43]

    Wiley Interdiscip Rev Rna 2022:e1749

    Vilimova M, Pfeffer S: Post - transcriptional regulation of polycistronic microRNAs. Wiley Interdiscip Rev Rna 2022:e1749

    work page 2022

  44. [44]

    Nature Biotechnology 2020, 38(6):675-678

    Goldman MJ, Craft B, Hastie M, Repečka K, McDade F, Kamath A, Banerjee A, Luo Y, Rogers D, Brooks AN et al: Visualizing and interpreting cancer genomics data via the Xena platform. Nature Biotechnology 2020, 38(6):675-678

    work page 2020

  45. [45]

    Nature 2020, 581(7808):303-309

    Han X, Zhou Z, Fei L, Sun H, Wang R, Chen Y, Chen H, Wang J, Tang H, Ge W et al: Construction of a human cell landscape at single-cell level. Nature 2020, 581(7808):303-309

    work page 2020

  46. [46]

    In: International conference on machine learning: 2015

    Ioffe S, Szegedy C: Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: International conference on machine learning: 2015. PMLR: 448-456

    work page 2015

  47. [47]

    J Mach Learn Res 2014, 15:1929-1958

    Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R: Dropout: A Simple Way to Prevent Neural Networks from Overfitting. J Mach Learn Res 2014, 15:1929-1958

    work page 2014

  48. [48]

    arXiv preprint arXiv:180308375 2018

    Agarap AF: Deep learning using rectified linear units (relu). arXiv preprint arXiv:180308375 2018

    work page 2018

  49. [49]

    Nucleic Acids Research 2020, 48(W1):W521-W528

    Kern F, Fehlmann T, Solomon J, Schwed L, Grammes N, Backes C, Van Keuren-Jensen K, Craig DW, Meese E, Keller A: miEAA 2.0: integrating multi-species microRNA enrichment analysis and workflow management systems. Nucleic Acids Research 2020, 48(W1):W521-W528

    work page 2020

  50. [50]

    Proceedings of the International AAAI Conference on Web and Social Media 2009, 3((1)):361-362

    Bastian M, Heymann S, Jacomy M: Gephi: An open source software for exploring and manipulating networks. Proceedings of the International AAAI Conference on Web and Social Media 2009, 3((1)):361-362

    work page 2009

  51. [51]

    The Gene Ontology Consortium

    Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT et al: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 2000, 25(1):25-29

    work page 2000

  52. [52]

    The Innovation 2021, 2(3)

    Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L et al: clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. The Innovation 2021, 2(3)

    work page 2021

  53. [53]

    Bioinformatics 2018, 34(4):713-715

    Wong NW, Chen Y, Chen S, Wang X: OncomiR: an online resource for exploring pan-cancer microRNA dysregulation. Bioinformatics 2018, 34(4):713-715

    work page 2018

  54. [54]

    Nucleic Acids Res 2019, 47(W1):W556-W560

    Tang Z, Kang B, Li C, Chen T, Zhang Z: GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res 2019, 47(W1):W556-W560

    work page 2019

  55. [55]

    Genome Biology 2014, 15(12):550

    Love MI, Huber W, Anders S: Moderated estimation of fold change and dispersion for RNA- seq data with DESeq2. Genome Biology 2014, 15(12):550

    work page 2014

  56. [56]

    Cell 2021, 184(13):3573-3587 e3529

    Hao Y, Hao S, Andersen-Nissen E, Mauck WM, 3rd, Zheng S, Butler A, Lee MJ, Wilk AJ, Darby C, Zager M et al: Integrated analysis of multimodal single-cell data. Cell 2021, 184(13):3573-3587 e3529. 34 Supplementary material Figure S1: Box plot for model performance by 3-fold cross-validation. (Blue and red) Training and test dataset average PCC of miRNA amo...

    work page 2021