Abstract
Circadian rhythms, which are the natural cycles that dictate various physiological processes over a 24-h period, have been increasingly recognized as important in the management and treatment of various human diseases. However, the lack of sufficient data and reliable analysis methods have been a major obstacle to understanding the bidirectional interaction between circadian variation and human health. We have developed CircaKB, a comprehensive knowledgebase of circadian genes across multiple species. CircaKB is the first knowledgebase that provides systematic annotations of the oscillatory patterns of gene expression at a genome-wide level for 15 representative species. Currently, CircaKB contains 226 time-course transcriptome datasets, covering a wide variety of tissues, organs, and cell lines. In addition, CircaKB integrates 12 computational models to facilitate reliable data analysis and identify oscillatory patterns and their variations in gene expression. CircaKB also offers powerful functionalities to its users, including easy search, fast browsing, strong visualization, and custom upload. We believe that CircaKB will be a valuable tool and resource for the circadian research community, contributing to the identification of new targets for disease prevention and treatment. We have made CircaKB freely accessible at https://cdsic.njau.edu.cn/CircaKB.
Graphical Abstract
Graphical Abstract.
Introduction
Circadian rhythms, which last about 24 h, regulate almost all physiological and behavioral processes, such as the sleep-wake cycle, blood pressure, and temperature (1). These rhythms are controlled by the suprachiasmatic nucleus (SCN), the primary circadian pacemaker (2,3). Neurons in the SCN contain intricate transcriptional and translational feedback loops that consist of core clock genes (4–6). These genes drive circadian oscillations of clock-controlled genes (CCGs) at the cellular level and regulate various cellular and systemic functions (7,8). Disturbances of the circadian rhythms have long been considered a risk factor for neurological diseases and cancers (9,10). However, recent findings suggest that circadian dysfunction may be a potential contributor to disease pathogenesis (11,12). Therefore, a deeper understanding of the interactions between the circadian disruption and disease development holds great potential for translating circadian research into clinical medicine.
Deciphering the variations in the oscillatory patterns of gene expression in disease development is a crucial step for uncovering the molecular mechanisms underlying circadian rhythms. With the exponential growth of public transcriptome and proteome datasets in recent years, rhythmic patterns of gene expression can be assessed by analyzing time-course data using computational tools (13). CircaDB, constructed in 2013 (14), is the earliest database of mammalian circadian gene expression profiles. It only collected 677 time-course experimental samples from mice and humans and provided an online browse for the circadian oscillation of each transcript's expression. CirGRDB, focusing on the expression patterns of disease-related circadian RNAs, collected more than 4936 genome-wide analyses associated with 37 human/mouse tissues or cell lines (15). CircadiOmics was developed as an annotation repository and analytic webserver for time series omics data (16). It contains transcriptomic, metabolic, and proteomic data from 20 tissues/organs of 11 different species. However, there are two obvious limitations in the above repositories. (i) All the online data analysis and visualization are determined by only 1–2 computational models. (ii) There is no online platform that provides differential rhythmicity analysis. In addition, a knowledgebase, CGDB, reported the phase and amplitude of 27964 circadian-related genes validated by experiments, involving 68 animals, 39 plants, and 41 fungi (17). Unfortunately, CGDB lacks information on circadian patterns and their alterations at the genome-wide scale.
Therefore, our team developed CircaKB, a comprehensive knowledgebase containing circadian genes across multiple species. It is designed to be an interactive platform of resources and annotations for circadian oscillation detection and differential rhythmicity identification (Figure 1). The current version of CircaKB integrates 226 time-course transcriptome datasets and 5577 measurements from multiple international public repositories, covering 54 tissues across 15 species. Notably, CircaKB offers a rich algorithm library for identifying oscillatory patterns of gene expression at a genome-wide scale. Specifically, CircaKB can not only detect circadian rhythms in the time-course data under a single experimental condition, but also estimate the differences in rhythms between two conditions. The platform also provides strong visualization power to help users better understand the results of their data analysis. Furthermore, CircaKB provides a user-friendly web interface that allows users to query and browse the oscillation trends of circadian genes of interest from existing datasets. Users can also upload their own data for real-time analysis. CircaKB is now freely accessible to all users without the need for login credentials at https://cdsic.njau.edu.cn/CircaKB.
Figure 1.
Overview of the CircaKB platform. (A) The whole framework of CircaKB; (B) Statistics of samples in CircaKB.
Materials and methods
Data collection and preprocessing
The CircaKB database contains 226 datasets and 5577 samples obtained from various public repositories such as GEO (18), GTEX (19), EBI (20), GEN (21), spanning 15 species. The transcriptome data in the database were generated using 9 different experimental strategies, which include Poly(A)-seq (22), Gene microarray (23), Total RNA-seq (24), scRNA-seq (25), ATAC-seq (26), Nascent-seq (24), Quant-seq (27), TRAP-seq (28) and miRNA-seq (29) (Table 1). If a dataset has been normalized by the original authors, it is directly analyzed and stored. All raw gene microarray data undergoes gene identifier (ID) mapping (30,31). For the raw read count data from Poly(A)-seq or Total-RNA-seq, we implement TMM (trimmed mean of M values) normalization using edgeR before gene mapping (32). Preprocessing of scRNA-seq data involves quality control, log-normalization, and gene mapping (33). Gene ID mapping is implemented by using g:Profiler (34). In each dataset, genes with missing values and less than four time points are removed.
Table 1.
Data summary of CircaKB. The number of datasets for each species is presented
| Species | Poly(A)-seq | Gene microarray | Total RNA-seq | scRNA-seq | ATAC-seq | Nascent-seq | Quant-seq | TRAP-seq | miRNA-seq |
|---|---|---|---|---|---|---|---|---|---|
| M. musculus | 33 | 28 | 15 | 2 | 1 | ||||
| P. anubis | 64 | ||||||||
| D. melanogaster | 2 | 7 | 20 | 1 | |||||
| H. sapiens | 3 | 16 | 6 | 1 | |||||
| D. rerio | 3 | 1 | |||||||
| G. gallus | 2 | ||||||||
| M. mulatta | 2 | ||||||||
| O. sativa | 1 | 4 | 1 | ||||||
| A. thaliana | 2 | 3 | 1 | ||||||
| S. tuberosum | 2 | ||||||||
| G. max | 1 | ||||||||
| O. aries | 1 | ||||||||
| T. aestivum | 1 | ||||||||
| Z. mays | 1 | ||||||||
| S. cerevisiae | 1 |
Database design and construction
We created three tables in MySQL 8.0 to store metadata for all collected datasets, website News, and visitor tracking. To speed up webpage loading time, we stored each processed gene expression matrix (.csv file) and its analyzed results (JSON files) as OSS (Object Storage Service (35)) objects. These OSS objects were deployed with MinIO 8.5.10 (https://min.io). In addition, we used Redis 5.0.7 (https://redis.io/) to log IP addresses in real time for tracking website visits.
Algorithm library construction
We incorporated 12 statistical models into this online platform to identify oscillatory patterns of gene expression at a genome-wide scale. The first seven models (JTK_CYCLE (36), Cosinor (37), ARSER (38), Lomb-scargle (39), RAIN (40), Fisher's G-test (41), and Robust G-test (42)) are designed for detecting circadian oscillation and providing parameters of rhythmic patterns, such as period, amplitude, acrophase, mesor, and P-values (Table 2). The remaining five models (diffCircadian (43), CircaCompare (44), LimoRhyde (45), HANOVA (46), and robust DODR (46)) are designed for analyzing differential rhythmicity, systematically assessing differences of rhythm in amplitude, phase, mesor, and fit (of sinusoidal wave) (Table 3). All these algorithms were created with R language and can be called by the Java codes at backend.
Table 2.
Summary of seven computational models for detecting circadian rhythm
| Period | Amplitude | Acrophase | Mesor | P-value | Q-value | Equal time intervals | |
|---|---|---|---|---|---|---|---|
| JTK_CYCLE |
|
|
|
|
|
|
|
| Consinor |
|
|
|
|
|
|
|
| ARSER |
|
|
|
|
|
|
|
| Lomb-scargle |
|
|
|
|
|||
| RAIN |
|
|
|
|
|||
| Fisher's G-test |
|
|
|||||
| Robust G-test |
|
|
JTK_CYCLE, ARSER and RAIN only accept the input data with equal time intervals. In addition to Fisher's G-test and Robust G-test, which can only feedback P-value and Q-value, the other five algorithms can also estimate circadian parameters, such as period, amplitude, acrophase or mesor.
Table 3.
Summary of five computational models for identifying difference in circadian rhythms
| Diff. amplitude | Diff. phase | Diff. mesor | Diff. fit | Diff. rhythmicity | |
|---|---|---|---|---|---|
| diffCircadian |
|
|
|
|
|
| CircaCompare |
|
|
|
||
| LimoRhyde |
|
||||
| HANOVA |
|
||||
| Robust DODR |
|
LimoRhyde, HANOVA and Robust DODR only evaluate whether there is a significant difference between two circadian rhythms. Both CircaCompare and diffCircadian can be used to detect the difference in amplitude, phase, or mesor. In addition, diffCircadian can identify the difference in data fitting to a sinusoidal wave.
Website design and implementation
The frontend of CircaKB is designed by using the Vue framework and provides a user-friendly interface to display the oscillatory pattern of each gene in any existing dataset. The backend system, developed with Java 8.0 and Spring Boot 2.5.9, processes requests from the frontend interface, retrieves the processed data and analyzes results from the MinIO database. CircaKB platform also supports new data uploading for online analysis. Data access is through MyBatis-Plus or minio-java to ensure data security, scalability, and efficiency. We use Docker 20.10 to deploy the website on our server cdsic.njau.edu.cn. All webpages were tested on the major browsers, including Google Chrome, Microsoft Edge, and Safari.
Data visualization
ECharts was adopted to provide state-of-the-art web visualizations (47). On the CircaKB homepage, four interactive pie charts dynamically present statistical information of the collected data using SVG renderer. Echart functions are used to generate scatter plots, line charts, and periodic curves for each gene.
Results
Design and organization of CircaKB
As a comprehensive knowledgebase, CircaKB provides an interactive platform for the resource and annotation of circadian genes (Figure 1A). Firstly, CircaKB contains a rich data resource with 226 time-course transcriptome datasets covering nine types of experimental strategies. Secondly, it includes 12 statistical models for identifying oscillatory patterns of gene expression at the genome-wide scale. Specifically, seven models are used for circadian oscillation detection, while the remaining five models are for differential rhythmicity analysis. Finally, CircaKB has a well-developed web-based system that allows users to browse statistics of the data and query annotations regarding the circadian pattern of any gene.
Overview of data resources in CircaKB
CircaKB offers a comprehensive data resource that can be used to study the oscillatory patterns of circadian genes of interest from various datasets. In the current version, CircaKB integrates 226 time-course transcriptome datasets with 5577 measurements across 54 tissues and 15 species from multiple international public repositories, such as Gene Expression Omnibus (GEO) (18), the European Bioinformatics Institute (EBI) database (20), and Gene Expression Nebulas (GEN) (21) (Figure 1B). The organisms covered by CircaKB include mammals, vertebrates, insects, plants, and microbes. These rich datasets enable researchers to investigate molecular rhythms across different organisms and relate them to both normal and disease physiology.
Browse CircaKB and search for annotation of circadian genes
The homepage of CircaKB is divided into four main sections: the navigation menu (navbar), an overview of CircaKB, a news section, and a sample statistics board. The navbar has seven functional entrances to help users navigate (Figure 2A). The sample statistics board uses pie charts and collapse components to display statistical information about the existing datasets in CircaKB, focusing on species, tissues, conditions, and experimental strategies (Figure 2A).
Figure 2.
The main interfaces of CircaKB. (A) Interacting with the pie charts on the home page to search the datasets of interests; (B) The browse page for circadian oscillation detection; (C) The browse page for differential rhythmicity analysis.
There are two ways for users to locate one dataset they are interested in. The first way is by interacting with the pie charts or collapse components on the homepage and generating a search query with four keywords (Figure 2A). The datasets that match the query will be listed at the bottom of the homepage. Then, users can determine whether they want to detect circadian oscillation on a dataset with only one experimental condition (Figure 2B), or identify differential rhythmicity between two conditions (Figure 2C). The other way is to directly click ‘Circadian Oscillation’ or ‘Differential rhythmicity’ on the navigation menu to start browsing the periodicity of gene expression. In this way, users need to further select the species and tissue, and decide which dataset they want to browse.
When working with a dataset with a single experimental condition, users can examine the rhythmic patterns of gene expression at the genome-wide scale once a computational model is selected (Figure 2B). Genes exhibiting significant oscillation trends will be sorted by p-value and listed in a drop-down menu. Selecting any gene on this menu will promptly generate a periodic curve in the right panel of the webpage. Furthermore, the table located at the bottom of the webpage displays the parameters of a rhythmic pattern estimated by all seven models, including period, amplitude, acrophase, mesor, P-value and Q-value.
In Figure 2C, we present the procedure for statistically estimating the differences in rhythms when comparing two conditions. This procedure can be used to detect differences in rhythmic characteristics such as amplitude, phase, mesor, and fit (of sinusoidal wave). Additionally, genes with differential rhythmicity will be ranked in an interactive drop-down list. Upon selecting a gene, users can immediately view two periodic curves in the right panel. Finally, a table will display the significance of differences in specific rhythmic characteristics between groups.
In particular, the ‘Search’ function on navbar enables users to quickly retrieve the circadian patterns of a gene in different species. The details can be referred to the application examples.
Download CircaKB datasets
All time-course datasets are available for individual download on the ‘Datasets’ page. Users can easily access the dataset list from the navigation bar and download the data they are interested in. Additionally, the significant genes identified by any computational model are available for download via the ‘Download this gene list’ icon for further analysis (Figure 2B, C).
Upload data for real-time analysis
Newly generated datasets can be submitted from the ‘Upload’ page for real-time analysis (Figure 3). Users should first upload their own data in the specified format (Figure 3A), select a suitable computational model (Figure 3B), and then view the analysis results immediately on the webpage (Figure 3C). Model selection depends on the characteristics of the data and the requirements of analysis. Tables 2 and 3 facilitate users in selecting the optimal algorithm for their specific analysis needs.
Figure 3.
A typical page for data uploading and online analysis. (A) Data uploading module; (B) Sample and model selection module; (C) Analysis result display module.
Knowledgebase applications
CircaKB is an important tool that systematically annotates oscillatory patterns of gene expression in time-course transcriptome data. To showcase the reliability of the resources and annotations provided by CircaKB, we selected three typical cases of mice and human to demonstrate the usefulness of this platform.
Case 1. Identify the circadian oscillation of gene expression in mouse liver
Previous studies have revealed that the liver circadian clock plays a significant role in preventing sleep disorders and diseases (48,49). Currently, CricaKB contains up to 30 time-course datasets of mouse liver. Here, we used mouse liver as an example to demonstrate how users can use CircaKB to implement circadian oscillation detection (Figure 4A). We selected three representative datasets of mouse liver with different sampling times (24 or 48 h) and sampling periods (1 or 2 h). Due to the differences in the underlying algorithms, the number of genes exhibiting significant oscillatory patterns screened by these models varies widely (Figure 4A(I)). It appears that JTK_CYCLE, Cosinor, and ARSER have excellent analysis capabilities and can provide almost all circadian parameters. In addition, we used the dataset GSE11923 to investigate whether there are significant differences in the analysis results of the above three models. As shown in Figure 4A(II), at least 85% of the genes between Cosinor and ARSER overlap. Most of the oscillatory genes detected by JTK_CYCLE can also be identified by Cosionr. Lastly, we compared the periodic curves of clock gene Arntl predicted by the aforementioned three models. Figure 4A(III) shows that the oscillatory patterns of Arntl inferred by these three models are similar.
Figure 4.
Two case studies of CircaKB’s applications for investigating circadian patterns on single dataset. (A) Identify the circadian oscillation of gene expression in mouse liver; (B) Identify alterations in the circadian patterns of gene expression between older and younger individuals.
Case 2. Identify changes in the circadian patterns of gene expression between older and younger individuals
A previous study reported that the aging process impacts the circadian rhythms of gene expression in the human brain (50). However, the effects of aging on molecular rhythms in the human brain are still poorly understood due to the lack of comprehensive data and reliable analyzing tools. In this study, we used the dataset GSE71620 as a representative case to show how CircaKB systematically identifies the differences in rhythms comparing two experimental conditions (Figure 4B). The data was generated from the BA11 brain region of young and old donors. Our analysis revealed that many genes exhibited differences in rhythms in the aged brain, including amplitude change, phase shift, base shift, and fit change (Figure 4B(I)). Particularly, we identified the AD-related biomarker BACE2 (51,52) with two types of variations in the oscillatory patterns, suggesting that BACE2 may be involved in regulating normal aging through circadian alterations (Figure 4B(II)).
Case 3. Investigate the circadian patterns of a gene of interest in different organisms
The core clock gene Cry1 was selected as a representative case. Users can start by navigating to the search page, where they input the gene name (Figure 5A). Upon clicking the ‘Search’ button, the search function was implemented to retrieve all matched datasets in CircaKB. The search results are then displayed in a table showing the organisms currently associated with Cry1 in CircaKB (Figure 5B). Selecting a specific tissue for an organism of interest and clicking the ‘View’ icon will lead to a detailed page for further exploration. Figure 5C shows the circadian patterns of Cry1 in mouse hearts predicted using datasets from different labs. The circadian patterns of Cry1 expression inferred from different data sources are very close: (i) the periods of all circadian curves fall within the range of 24–27 h; (ii) the peaks occur between 17.6 and 21.7 h. Additionally, we examined the circadian profiles of CRY1 in the same tissue across three species. Supplementary Figure S1 indicates that the circadian rhythms of CRY1 in retina are significantly different among G. gallus, R. macaque, and O. baboon.
Figure 5.
An application example using different data sources to investigate the circadian patterns of a specific gene. (A) Search page in CircaKB; (B) Search results for the core clock gene Cry1 involve different organisms; (C) Circadian patterns of Cry1 in mouse hearts predicted using datasets from different labs.
Discussion and future development
CircaKB is the first and only knowledgebase that can systematically annotate oscillatory patterns of gene expression in time-course transcriptome data across multiple species. It can be routinely used in basic research to understand the molecular significance of oscillatory patterns and discover new targets for disease prevention and treatment. We believe that CircaKB will be an invaluable resource for the circadian research community that could power many future studies.
CircaKB is an advanced platform that centralizes data resources crucial to current circadian research. This platform currently hosts an extensive collection of representative species, including mammals, vertebrates, insects, plants, and microbes. It contains diverse functional modules that provide users with reliable analysis and multiple visualization options. CircaKB offers fast webpage response time for accessing annotations. In the test with an upload bandwidth of 30 Mbps, download bandwidth of 100Mbps, and an average latency of 18ms, we found that the average loading times for the ‘Circadian oscillation’ and ‘Differential Rhythmicity’ webpages were only 645.66 and 1548.24 ms, respectively.
In chronobiology, ultradian rhythms have shorter periods and higher frequencies compared to circadian rhythms (53,54). These rhythms include various physiological processes, such as sleep stages, blood circulation, and heart rate (55,56). In this study, we included an RNA-seq dataset, GSE220120, generated by Zhu et al., which examined transcriptional rhythms in human white blood cells from three healthy donors and identified robust 
12 h transcriptional rhythms (57). Supplementary Figure S2 demonstrates that CircaKB effectively detects ultradian rhythms of gene expression.
We plan to continuously enhance CircaKB’s circadian research capabilities by adding more content and features. First, we will expand our database by collecting relevant data for existing species and other species, and will include additional computational pipelines to provide reliable annotations. Notably, we also developed a new model to decipher oscillatory patterns of gene expression using untimed transcriptome datasets, which will allow us to integrate public repositories without temporal information, such as TCGA, into CircaKB. Overall, we believe that CircaKB will have a significant impact on both basic circadian research and clinical medicine.
Supplementary Material
Acknowledgements
The authors gratefully acknowledge the Bioinformatics Center, Nanjing Agricultural University for the High-Performance Computing platform (NJAU-HPC). We also thank Dr. Weiling Zhao at UNC Chapel Hill for her valuable advice in writing. Additionally, we thank all the original researchers who selflessly shared their data and codes.
Author contributions: Xingchen Zhu: Data collection and analysis, Software, Visualization, Writing; Xiao Han: Conceptualization, Data analysis; Zhijin Li: Data analysis; Xiaobo Zhou: Conceptualization, Review & editing; Seung-Hee Yoo: Conceptualization; Zheng Chen: Resource; Zhiwei Ji: Supervision, Project administration, Funding acquisition, Conceptualization, Methodology, Review & editing.
Contributor Information
Xingchen Zhu, College of Artificial Intelligence, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China; Center for Data Science and Intelligent Computing, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China.
Xiao Han, College of Artificial Intelligence, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China; Center for Data Science and Intelligent Computing, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China.
Zhijin Li, Department of Neurosurgery, The First Affiliated Hospital of USTC (Anhui Provincial Hospital), Division of Life Science and Medicine, University of Science and Technology of China, Hefei, Anhui 230036, China.
Xiaobo Zhou, School of Biomedical Informatics, The University of Texas Health Science Center at Houston, 7000 Fannin Street, Houston, TX 77030, USA.
Seung-Hee Yoo, Department of Biochemistry and Molecular Biology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX 77030, USA.
Zheng Chen, Department of Biochemistry and Molecular Biology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX 77030, USA.
Zhiwei Ji, College of Artificial Intelligence, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China; Center for Data Science and Intelligent Computing, Nanjing Agricultural University, No. 1 Weigang Rd., Nanjing, Jiangsu 210095, China.
Data availability
The CircaKB database is freely accessible to all academic users at http://cdsic.njau.edu.cn/CircaKB. Each processed dataset can be downloaded from the ‘Datasets’ page. Additionally, the researchers can also download the data from Zenodo at https://doi.org/10.5281/zenodo.12741531. Significant genes selected by a specific model on any specific dataset can be exported by clicking the icon ‘Download this gene list’. The codes of CircaKB’s algorithm library are deposited at https://doi.org/10.6084/m9.figshare.26386285.
Supplementary data
Supplementary Data are available at NAR Online.
Funding
Fundamental Research Funds for the Central Universities [YDZX2024009]; Agricultural Science and Technology Innovation Foundation of Jiangsu Province [CX (23) 3125]; The startup award of new professors at Nanjing Agricultural University [106/804001]. Funding for open access charge: Agricultural Science and Technology Innovation Foundation of Jiangsu Province [CX (23) 3125].
Conflict of interest statement. None declared.
References
- 1. Bell-Pedersen D., Cassone V.M., Earnest D.J., Golden S.S., Hardin P.E., Thomas T.L., Zoran M.J.. Circadian rhythms from multiple oscillators: lessons from diverse organisms. Nat. Rev. Genet. 2005; 6:544–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Hastings M.H., Maywood E.S., Brancaccio M.. Generation of circadian rhythms in the suprachiasmatic nucleus. Nat. Rev. Neurosci. 2018; 19:453–469. [DOI] [PubMed] [Google Scholar]
- 3. Hastings M.H., Brancaccio M., Maywood E.S.. Circadian pacemaking in cells and circuits of the suprachiasmatic nucleus. J. Neuroendocrinol. 2014; 26:2–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Jones J.R., Chaturvedi S., Granados-Fuentes D., Herzog E.D.. Circadian neurons in the paraventricular nucleus entrain and sustain daily rhythms in glucocorticoids. Nat. Commun. 2021; 12:5763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Herzog E.D., Hermanstyne T., Smyllie N.J., Hastings M.H.. Regulating the suprachiasmatic nucleus (SCN) circadian clockwork: interplay between cell-autonomous and circuit-level mechanisms. Cold Spring Harb. Perspect. Biol. 2017; 9:a027706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Duncan M.J. Interacting influences of aging and Alzheimer's disease on circadian rhythms. Eur. J. Neurosci. 2020; 51:310–325. [DOI] [PubMed] [Google Scholar]
- 7. O’Neill J.S., Feeney K.A.. Circadian redox and metabolic oscillations in mammalian systems. Antioxid. Redox. Signal. 2014; 20:2966–2981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. O’Neill J.S., Reddy A.B.. The essential role of cAMP/Ca2+ signalling in mammalian circadian timekeeping. Biochem. Soc. Trans. 2012; 40:44–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Leng Y., Musiek E.S., Hu K., Cappuccio F.P., Yaffe K.. Association between circadian rhythms and neurodegenerative diseases. Lancet Neurol. 2019; 18:307–318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Nassan M., Videnovic A.. Circadian rhythms in neurodegenerative disorders. Nat. Rev. Neurol. 2022; 18:7–24. [DOI] [PubMed] [Google Scholar]
- 11. Samanta S., Ali S.. Impact of circadian clock dysfunction on human health. Explor. Neurosci. 2022; 1:4–30. [Google Scholar]
- 12. Ruan W., Yuan X., Eltzschig H.K.. Circadian rhythm as a therapeutic target. Nat. Rev. Drug Discov. 2021; 20:287–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Anafi R.C., Francey L.J., Hogenesch J.B., Kim J.. CYCLOPS reveals human transcriptional rhythms in health and disease. Proc. Natl. Acad. Sci. U.S.A. 2017; 114:5312–5317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Pizarro A., Hayer K., Lahens N.F., Hogenesch J.B.. CircaDB: a database of mammalian circadian gene expression profiles. Nucleic Acids Res. 2013; 41:D1009–D1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Li X., Shi L., Zhang K., Wei W., Liu Q., Mao F., Li J., Cai W., Chen H., Teng H.et al.. CirGRDB: a database for the genome-wide deciphering circadian genes and regulators. Nucleic Acids Res. 2018; 46:D64–D70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Samad M., Agostinelli F., Sato T., Shimaji K., Baldi P.. CircadiOmics: circadian omic web portal. Nucleic Acids Res. 2022; 50:W183–W190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Li S., Shui K., Zhang Y., Lv Y., Deng W., Ullah S., Zhang L., Xue Y.. CGDB: a database of circadian genes in eukaryotes. Nucleic Acids Res. 2017; 45:D397–D403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Edgar R., Domrachev M., Lash A.E.. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 2002; 30:207–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. The GTEx Consortium The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science. 2020; 369:1318–1330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Emmert D.B., Stoehr P.J., Stoesser G., Cameron G.N.. The European Bioinformatics Institute (EBI) databases. Nucleic Acids Res. 1994; 22:3445–3449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Zhang Y., Zou D., Zhu T., Xu T., Chen M., Niu G., Zong W., Pan R., Jing W., Sang J.et al.. Gene Expression Nebulas (GEN): a comprehensive data portal integrating transcriptomic profiles across multiple species at both bulk and single-cell levels. Nucleic Acids Res. 2022; 50:D1016–D1024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Yu F., Zhang Y., Cheng C., Wang W., Zhou Z., Rang W., Yu H., Wei Y., Wu Q., Zhang Y.. Poly(A)-seq: a method for direct sequencing and analysis of the transcriptomic poly(A)-tails. PLoS One. 2020; 15:e0234696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Watson A., Mazumder A., Stewart M., Balasubramanian S.. Technology for microarray analysis of gene expression. Curr. Opin. Biotechnol. 1998; 9:609–614. [DOI] [PubMed] [Google Scholar]
- 24. Ameur A., Zaghlool A., Halvardson J., Wetterbom A., Gyllensten U., Cavelier L., Feuk L.. Total RNA sequencing reveals nascent transcription and widespread co-transcriptional splicing in the human brain. Nat. Struct. Mol. Biol. 2011; 18:1435–1440. [DOI] [PubMed] [Google Scholar]
- 25. Saliba A.E., Westermann A.J., Gorski S.A., Vogel J.. Single-cell RNA-seq: advances and future challenges. Nucleic Acids Res. 2014; 42:8845–8860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Grandi F.C., Modi H., Kampman L., Corces M.R.. Chromatin accessibility profiling by ATAC-seq. Nat. Protoc. 2022; 17:1518–1552. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Herzog V.A., Reichholf B., Neumann T., Rescheneder P., Bhat P., Burkard T.R., Wlotzka W., von Haeseler A., Zuber J., Ameres S.L.. Thiol-linked alkylation of RNA to assess expression dynamics. Nat. Methods. 2017; 14:1198–1204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Mayor-Ruiz C., Dominguez O., Fernandez-Capetillo O.. Trap(Seq): an RNA sequencing-based pipeline for the identification of gene-Trap insertions in mammalian cells. J. Mol. Biol. 2017; 429:2780–2789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Tam S., Tsao M.S., McPherson J.D.. Optimization of miRNA-seq data preprocessing. Brief Bioinform. 2015; 16:950–963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Del Olmo M., Sporl F., Korge S., Jurchott K., Felten M., Grudziecki A., de Zeeuw J., Nowozin C., Reuter H., Blatt T.et al.. Inter-layer and inter-subject variability of diurnal gene expression in human skin. NAR Genom Bioinform. 2022; 4:lqac097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Federico A., Saarimaki L.A., Serra A., Del Giudice G., Kinaret P.A.S., Scala G., Greco D.. Microarray data preprocessing: from experimental design to differential analysis. Methods Mol. Biol. 2022; 2401:79–100. [DOI] [PubMed] [Google Scholar]
- 32. Zhao Y., Li M.C., Konate M.M., Chen L., Das B., Karlovich C., Williams P.M., Evrard Y.A., Doroshow J.H., McShane L.M.. TPM, FPKM, or normalized counts? A comparative study of quantification measures for the analysis of RNA-seq data from the NCI patient-derived models repository. J. Transl. Med. 2021; 19:269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Wang Z., Xie X., Liu S., Ji Z.. scFseCluster: a feature selection-enhanced clustering for single-cell RNA-seq data. Life Sci Alliance. 2023; 6:e202302103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Kolberg L., Raudvere U., Kuzmin I., Adler P., Vilo J., Peterson H.. g:profiler-interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Res. 2023; 51:W207–W212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Sivaji A., Tzuaan S.. Website user experience (UX) testing tool development using Open Source Software (OSS). 2012 Southeast Asian Network of Ergonomics Societies Conference (SEANES). 2012; 1–6. [Google Scholar]
- 36. Hughes M.E., Hogenesch J.B., Kornacker K.. JTK_CYCLE: an efficient nonparametric algorithm for detecting rhythmic components in genome-scale data sets. J. Biol. Rhythms. 2010; 25:372–380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Cornelissen G. Cosinor-based rhythmometry. Theor. Biol. Med. Model. 2014; 11:16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Yang R., Su Z.. Analyzing circadian expression data by harmonic regression based on autoregressive spectral estimation. Bioinformatics. 2010; 26:i168–i174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Glynn E.F., Chen J., Mushegian A.R.. Detecting periodic patterns in unevenly spaced gene expression time series using Lomb-Scargle periodograms. Bioinformatics. 2006; 22:310–316. [DOI] [PubMed] [Google Scholar]
- 40. Thaben P.F., Westermark P.O.. Detecting rhythms in time series with RAIN. J. Biol. Rhythms. 2014; 29:391–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Wichert S., Fokianos K., Strimmer K.. Identifying periodically expressed transcripts in microarray time series data. Bioinformatics. 2004; 20:5–20. [DOI] [PubMed] [Google Scholar]
- 42. Ahdesmaki M., Lahdesmaki H., Pearson R., Huttunen H., Yli-Harja O.. Robust detection of periodic time series measured from biological systems. BMC Bioinf. 2005; 6:117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Ding H., Meng L., Liu A.C., Gumz M.L., Bryant A.J., McClung C.A., Tseng G.C., Esser K.A., Huo Z.. Likelihood-based tests for detecting circadian rhythmicity and differential circadian patterns in transcriptomic applications. Brief Bioinform. 2021; 22:bbab224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Parsons R., Parsons R., Garner N., Oster H., Rawashdeh O.. CircaCompare: a method to estimate and statistically support differences in mesor, amplitude and phase, between circadian rhythms. Bioinformatics. 2020; 36:1208–1212. [DOI] [PubMed] [Google Scholar]
- 45. Singer J.M., Hughey J.J.. LimoRhyde: a flexible approach for differential analysis of rhythmic transcriptome data. J. Biol. Rhythms. 2019; 34:5–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Thaben P.F., Westermark P.O.. Differential rhythmicity: detecting altered rhythmicity in biological data. Bioinformatics. 2016; 32:2800–2808. [DOI] [PubMed] [Google Scholar]
- 47. Li D., Mei H., Shen Y., Su S., Zhang W., Wang J., Zu M., Chen W.. ECharts: a declarative framework for rapid construction of web-based visualization. Visual Informatics. 2018; 2:136–146. [Google Scholar]
- 48. Bolshette N., Ibrahim H., Reinke H., Asher G.. Circadian regulation of liver function: from molecular mechanisms to disease pathophysiology. Nat. Rev. Gastroenterol. Hepatol. 2023; 20:695–707. [DOI] [PubMed] [Google Scholar]
- 49. Tahara Y., Shibata S.. Circadian rhythms of liver physiology and disease: experimental and clinical evidence. Nat. Rev. Gastroenterol. Hepatol. 2016; 13:217–226. [DOI] [PubMed] [Google Scholar]
- 50. Li J.Z., Bunney B.G., Meng F., Hagenauer M.H., Walsh D.M., Vawter M.P., Evans S.J., Choudary P.V., Cartagena P., Barchas J.D.et al.. Circadian patterns of gene expression in the human brain and disruption in major depressive disorder. Proc. Natl. Acad. Sci. U.S.A. 2013; 110:9950–9955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Yeap Y.J., Kandiah N., Nizetic D., Lim K.L.. BACE2: a promising neuroprotective candidate for Alzheimer's disease. J. Alzheimers Dis. 2023; 94:S159–S171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Farzan M., Schnitzler C.E., Vasilieva N., Leung D., Choe H.. BACE2, a beta -secretase homolog, cleaves at the beta site and within the amyloid-beta region of the amyloid-beta precursor protein. Proc. Natl. Acad. Sci. U.S.A. 2000; 97:9712–9717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Coskun A., Zarepour A., Zarrabi A.. Physiological rhythms and biological variation of biomolecules: the road to personalized laboratory medicine. Int. J. Mol. Sci. 2023; 24:6275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Smolensky M.H., Haus E.. Circadian rhythms and clinical medicine with applications to hypertension. Am. J. Hypertens. 2001; 14:280S–290S. [DOI] [PubMed] [Google Scholar]
- 55. Cajochen C., Reichert C.F., Munch M., Gabel V., Stefani O., Chellappa S.L., Schmidt C.. Ultradian sleep cycles: frequency, duration, and associations with individual and environmental factors-A retrospective study. Sleep Health. 2024; 10:S52–S62. [DOI] [PubMed] [Google Scholar]
- 56. Shannahoff-Khalsa D.S., Kennedy B., Yates F.E., Ziegler M.G.. Ultradian rhythms of autonomic, cardiovascular, and neuroendocrine systems are related in humans. Am. J. Physiol. 1996; 270:R873–R887. [DOI] [PubMed] [Google Scholar]
- 57. Zhu B., Liu S., David N.L., Dion W., Doshi N.K., Siegel L.B., Amorim T., Andrews R.E., Naveen Kumar G.V., Li H.et al.. Evidence for conservation of primordial ∼12-hour ultradian gene programs in humans under free-living conditions. 2023; bioRxiv doi:23 December 2023, preprint: not peer reviewed 10.1101/2023.05.02.539021. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The CircaKB database is freely accessible to all academic users at http://cdsic.njau.edu.cn/CircaKB. Each processed dataset can be downloaded from the ‘Datasets’ page. Additionally, the researchers can also download the data from Zenodo at https://doi.org/10.5281/zenodo.12741531. Significant genes selected by a specific model on any specific dataset can be exported by clicking the icon ‘Download this gene list’. The codes of CircaKB’s algorithm library are deposited at https://doi.org/10.6084/m9.figshare.26386285.