Understanding the Molecular Landscape of Alopecia Areata: Key Genes, Regulatory Networks, and Therapeutic Targets Authors: Ahmed Ibrahim AbdElneam, Mohammed Saleh Al‐Dhubaibi, Saleh Salem Bahaj, Ghada Farouk Mohammed, Lina Mohammed Atef Published: 2026 ABSTRACT ABSTRACTIntroductionAlopecia areata (AA) is an autoimmune disorder characterized by hair loss due to immune‐mediated destruction of hair follicles. The polygenic nature of AA, influenced by genetic and environmental factors, complicates its pathogenesis. Advances in bioinformatics have facilitated understanding of AA's genetic basis, providing insights into key molecular pathways and regulatory mechanisms.Materials and MethodsThis study utilized the GEO database (dataset GSE111061) to analyze differentially expressed genes (DEGs) between healthy controls and AA patients. Functional enrichment and protein‐protein interaction (PPI) network analyzes identified hub genes, transcription factors (TFs), and miRNAs. Tools such as STRING, Cytoscape, and KEGG were employed to uncover key interactions and pathways.ResultsAn aggregate of 5134 DEGs was discovered, with 2576 elevated genes constituting a PPI network. Key pathways included chemokine signaling and cytokine‐cytokine receptor interactions. Ten hub genes (e.g., CCL5, CCR7, and CXCL9) and top TFs (ZNF683, FOXP3, and TBX21) were identified. Additionally, 49 miRNAs were predicted to regulate AA‐related genes.ConclusionsThis work focuses on critical molecular regulators involved in AA pathogenesis, such as TFs and miRNAs. These findings suggest possible treatment targets and shed light on the molecular mechanisms of AA.Keywords: Alopecia areata, bioinformatics, cytokine signaling, miRNAs, therapeutic targets, transcription factors Introduction Alopecia areata (AA) is an autoimmune condition characterized by hair loss, where the immune system mistakenly attacks hair follicles [1, 2]. It often presents with cycles of relapse and remission, making management challenging [1, 3, 4]. Approximately 2% of the population is affected by AA during their lifetime [5, 6, 7, 8]. Evidence suggests a strong genetic component, with 0%–8.6% of adult patients and 10%–51.6% of children having a family history of AA [9, 10]. Although statistics on youngsters suggest that rates of genetic connection range from 10% to 51.6% [11, 12]. Males are more likely than females to have a familial association, and the occurrence among siblings, twins, and multi‐generational families further highlights its heritability [13, 14]. Early genetic studies focused on candidate gene associations, particularly within the HLA‐D region on chromosome 6, known for regulating susceptibility to AA [15]. The National AA Registry Repository permitted family‐centered linkage studies and GWAS analysis, which validated the polygenic nature of AA [1, 16]. Bioinformatics, combining biology, computer science, and data analysis, plays a critical role in understanding complex conditions like AA [17]. It has enabled advancements like genome mapping and personalized medicine through next‐generation sequencing and reverse vaccinology [18]. These innovations have transformed disease research, vaccine development, and targeted therapies, offering insights into AA's genetic underpinnings [19]. The Aim of the Research Our research intends to uncover key genes and regulatory pathways involved in the emergence of AA through the analysis of microarray data sets from the GEO database with bioinformatics techniques. We are interested in studying the role of miRNAs and transcription factors in regulating AA‐related biological pathways. This strategy aims to identify novel treatment targets and enhance AA management. Methods Microarray Activity Profiling in GEO The study, approved by the ethical committee (ERC_SU_F_202300031), utilized the GEO database (http://www.ncbi.nlm.nih.gov/geo/) to identify gene expression profiling studies on AA. A comprehensive search using keywords like “AA” filtered for “Homo sapiens,” priority was given to studies with larger sample sizes and clearly defined case‐control groupings. Studies involving animal models, cell lines, or single‐sample data were excluded. One dataset meeting inclusion criteria. The selected dataset, GSE111061 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE111061), was derived from the GPL16791 Illumina HiSeq. 2500 platform, contained mRNA transcriptome data from six healthy controls and six AA patients [20]. Analysis of Microarray Set GSE111061 The GSE111061 dataset (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE111061) Differential gene expression (DEGs) analysis was carried out using GEO2R, a web‐based tool that leverages the limma package in R, dividing samples into healthy and patient groups. Statistical significance was assessed using the Benjamini‐Hochberg method to control the false discovery rate (FDR). Genes meeting the criteria of an adjusted p‐value below 0.05. The severity of AA included in this study is moderate and severe only. The analysis focused on upregulated genes, as they reveal activation pathways, serve as potential biomarkers, and correlate with disease progression, aligning with the activation‐driven mechanisms underlying AA. Network and Module Analysis for Protein‐Protein Interactions (PPI) In this study, gene interactions among differentially expressed genes (DEGs) were predicted using STRING v10.5 (http://string-db.org) [21]. PPI network was constructed with Cytoscape v3.7.1, and key proteins were ranked using the CytoHubba plugin. We add a group of criteria in the selection of PPI by excluding any unreacted proteins and utilizing an FDR threshold set at 0.07. Analyzing Biological, Cellular, Molecular Processes, and KEGG Pathways Utilizing Shiny GO 0.80 (http://bioinformatics.sdstate.edu/go/), Amigo (https://amigo.geneontology.org/amigo), and KEGG (https://www.genome.jp/kegg/pathway.html) with FDR < 0.05, we analyzed biological, cellular, and molecular pathways, uncovering complex interactions and mechanisms with enhanced insights [22]. Selection and Prediction of Appropriate TFs TFs were predicted using ChEA3 (https://maayanlab.cloud/chea3/#top) by submitting the top 10 genes. Cytoscape facilitated molecular network construction, identifying 10 high‐ranking TFs from integrated database analyzes. Analysis and Prediction of Appropriate miRNAs for Genes Using MiRTarBase 9.0 (https://mirtarbase.cuhk.edu.cn/), TargetScan (https://www.targetscan.org/vert_80/), miRDB (https://mirdb.org/), and miRWalk (http://mirwalk.umm.uni-heidelberg.de/) with 95% CI, potential miRNA‐target gene interactions were predicted. Cytoscape facilitated detailed molecular network analysis. The specific criteria that encompassed a variety of factors, including, among others, a high binding probability indicative of strong gene interaction, which is quantified as equaling [1], low binding energy that suggests favorable thermodynamic conditions quantified as equal to or lower than (−16), and the essential presence of the distinguished AU ‐rich elements AREs that are greater than or equal to (0.60), which are well recognized for their significant roles in post‐transcriptional regulation. Results DEGs for Integrated Analysis of Microarray Datasets A comprehensive examination of 20,447 genes was conducted, which culminated in the discovery of 5134 DEGs, comprising 2558 that were downregulated and 2576 that exhibited upregulation (Figure 1A,B). (A) Volcano plots showing the upregulated and downregulated genes in GSE111061 (The volcano plots shown identify genes that are p < 0.05 and are used as criteria for significance). (B) Principal component analysis (PCA) map of DEGs. The data was analyzed by GEO2R on the NCBI website. PPI Network Construction for Upregulated Genes A total of 519 upregulated genes were identified, forming an interaction network with 394 nodes and 419 edges. The top 10 genes, including CCL5, CCR7, and CXCL9, were analyzed using Cytoscape and CytoHubba (Table 1). The network's diameter was 2, and the overall node grade was 2.13. (Figure 2). Table 1Top 10 genes ranking and score.RankingTop 10 genesScore1CCL5167,7132CCR7166,1823CCR5165,1464CXCL9163,6245CXCR3163,4406CXCL10162,8607CXCR6122,4728CD8A92,3659CCL481,36010CCL1380,640Abbreviations: CCL5, C‐C motif chemokine ligand 5; CCR7, C‐C motif chemokine receptor 7; CCR5, C‐C motif chemokine receptor 5; CXCL9, C‐X‐C motif chemokine ligand 9; CXCR3, C‐X‐C motif chemokine receptor 3; CXCL10, C‐X‐C motif chemokine ligand 10; CXCR6, C‐X‐C motif chemokine receptor 6; CD8A, CD8 subunit alpha; CCL4, C‐C motif chemokine ligand 4; CCL13,C‐C motif chemokine ligand 13. Top 10 in network string interactions by MCC method. Top 10 genes ranking and score. Abbreviations: CCL5, C‐C motif chemokine ligand 5; CCR7, C‐C motif chemokine receptor 7; CCR5, C‐C motif chemokine receptor 5; CXCL9, C‐X‐C motif chemokine ligand 9; CXCR3, C‐X‐C motif chemokine receptor 3; CXCL10, C‐X‐C motif chemokine ligand 10; CXCR6, C‐X‐C motif chemokine receptor 6; CD8A, CD8 subunit alpha; CCL4, C‐C motif chemokine ligand 4; CCL13,C‐C motif chemokine ligand 13. Top 10 in network string interactions by MCC method. The PPI network of the top 10 DEGs in AA. The red color indicates the highest degree, and yellow indicates the lowest degree. The analysis was done by Cytoscape software and CytoHubba tool. Functional Enrichment Assessment for the Top 10 Genes The top biological pathways included chemokine‐mediated activation and responses to chemokines (FDR = 3.8941E‐18), involving genes like CXCL9, CXCL10, and CCL5 (Figure 3A). Cellular processes linked to plasma membrane components showed significant enrichment (FDR = 0.0229), with genes like CCR7 and CD8A (Figure 3B). Molecular processes highlighted chemokine activity and receptor binding (FDR = 3.599E‐09), involving CXCL9 and CCL13 (Figure 3C). KEGG analysis found amplification in chemokine communication and cytokine‐cytokine receptor coupling (FDR = 7.72E‐12) with genes like CCR7 and CXCR6 (Figures 3, 4, 5). Gene Ontology functional enrichment and KEGG pathways of DEGs (FDR < 0.05). (A) Biological process. (B) Cellular components. (C) Molecular functions. (D) KEGG pathway. FDR, false discovery rate. The charts are designed by the Shiny GO 0.80 tool. Chemokine signaling KEGG pathway. The red color indicates the gene's effect in the pathway from this study. The diagram is designed by the Shiny GO 0.80 tool and the KEGG pathway website. Cytokine ‐cytokine receptor interaction KEGG pathway. The red color indicates gene effects in the pathway from the top 10 genes in this study. The diagram was designed by the Shiny GO 0.80 tool and the KEGG pathway website. TFs Analysis and Prediction for the Top 10 Selected Genes The TFs were identified using ChEA3, identifying the top 10 based on their rank (Figure 6). The regulatory network and GO analysis are shown in (Figure 7A,B). The network, with 20 nodes and 54 edges, is visualized in (Figure 8A,B), revealing key interactions. The cluster heatmap (Figure 8C) shows shared genes among TFs. (Table 2) highlights TFs regulating the top nine genes, with FOXP3, ZNF683, and TBX21 controlling nine genes, and BATF and PRDM1 regulating six genes. Bar charts corresponding to the top TFs of the selected library display the ‐log10 (p‐value) for ChEA TF target library results and an integrated rank score for the integrated library results. The top 10 TFs are on the y‐axis. Global network analysis representation of top‐ranked TFs displaying putative interactions according to tissue‐specific, based on averaged integrated ranks across multiple libraries and databases (ChEA3 libraries) Based on the TCGFA TF network (The blue color indicates the top 10 TFs) Based on the GTEx network. (A) Local network analysis representation of top‐ranked TFs displaying putative interactions, based on averaged integrated ranks across multiple libraries and databases (ChEA3 libraries). (B) TFs and genes regulatory network using Cytoscape software. The yellow color indicates genes, and the blue color indicates TFs. (C) Heat map cluster analysis showing relationships between genes and TFs the red color indicates several TFs share in one gene. The analysis was done by ChEA3 libraries. Table 2TFs and overlapping genes.TFOverlapping genesIntegrated scale rankingLibraryBATF (Basic leucine zipper transcription factor, ATF‐like)CD8A, CCL5, CXCR3, CCL4, CCR7, CCR56.143E‐4* ARCHS4 CoexpressionTCF7 (Transcription factor 7 (T‐cell specific, HMG‐box)CD8A, CXCR3, CCR7, CCR56.223E‐4* GTExCoexpressionZNF683 (Zinc finger protein 683)CXCL10, CXCL9, CD8A, CCL5, CXCR3, CCL4, CCR7, CXCR6, CCR57.123E‐4* Enrichr queriesPRDM1 (PR domain containing 1, with ZNF domain)CD8A, CCL5, CXCR3, CCL4, CXCR6, CCR50.001229* ARCHS4 CoexpressionNFATC2 (Nuclear factor of activated T‐cells, cytoplasmic, calcineurin‐dependent 2)CD8A, CXCR3, CCR50.001245* GTExCoexpressionFOXP3 (Forkhead box P3)CXCL10, CXCL9, CD8A, CCL5, CXCR3, CCL4, CCR7, CXCR6, CCR50.001425* Enrichr queriesSCML4 (Sex comb on midleg‐like 4)CD8A, CCL5, CXCR30.001843* ARCHS4 CoexpressionZNF831 (Zinc finger protein 831)CD8A, CXCR3, CCR50.001867* GTExCoexpressionTBX21 (T‐box 21)CXCL10, CXCL9, CD8A, CCL5, CXCR3, CCL4, CCR7, CXCR6, CCR50.002137* Enrichr queriesTRAFD1 (TRAF‐type zinc finger domain containing 1)CXCL10, CCR7, CCR50.002457* ARCHS4 CoexpressionAbbreviations: CCL5, C‐C motif chemokine ligand 5; CCR7, C‐C motif chemokine receptor 7; CCR5, C‐C motif chemokine receptor 5; CXCL9, C‐X‐C motif chemokine ligand 9; CXCR3, C‐X‐C motif chemokine receptor 3; CXCL10, C‐X‐C motif chemokine ligand 10; CXCR6, C‐X‐C motif chemokine receptor 6; CD8A, CD8 subunit alpha; CCL4, C‐C motif chemokine ligand 4; CCL13,C‐C motif chemokine ligand 13.*Highly significant differences, p ≤ 0.05. TFs and overlapping genes. Abbreviations: CCL5, C‐C motif chemokine ligand 5; CCR7, C‐C motif chemokine receptor 7; CCR5, C‐C motif chemokine receptor 5; CXCL9, C‐X‐C motif chemokine ligand 9; CXCR3, C‐X‐C motif chemokine receptor 3; CXCL10, C‐X‐C motif chemokine ligand 10; CXCR6, C‐X‐C motif chemokine receptor 6; CD8A, CD8 subunit alpha; CCL4, C‐C motif chemokine ligand 4; CCL13,C‐C motif chemokine ligand 13. Highly significant differences, p ≤ 0.05. miRNA Analysis and Prediction of the Top 10 Selected Genes A total of 49 miRNAs were identified for regulating seven genes linked to AA, including 13 for CCR5 and 10 for CXCL9 (Figure 9). The network, constructed using Cytoscape, comprised 56 nodes, 49 edges, and exhibited low clustering with a density of 0.143. miRNAs and genes regulatory network for seven genes using Cytoscape software. The yellow color indicates genes, and the blue color indicates miRNAs. Discussion Bioinformatics has played a pivotal role in understanding complex polygenic diseases, such as AA, by identifying various genes linked to the disease and providing unique insights into its growth. In this study, 5134 DEGs were identified, including 2576 upregulated and 2558 downregulated genes. Among the upregulated genes, the top 10 hub genes included CCL5, CCL13, CCL4, CXCL9, CXCL10, CCR5, CCR7, CD8A, CXCR3, and CXCR6. These genes were identified as key players in the molecular pathways related to AA. The PPI network analysis revealed that genes like CCL5, CCR7, and CCR5 had higher interaction degrees, highlighting their critical roles in AA's pathogenesis. CCL5, for instance, exhibited a remarkable degree of 167,713, indicating its central role in immune signaling in AA. Previous studies support the significant involvement of CCL5 and CCR5 in AA. For example, Ito and colleagues highlighted CCR5 as a potential therapeutic target for AA [23]. Similarly, Subramanya et al. observed the elevation of CCL5 in skin of AA and Fraticelli et al. confirmed elevated cytokine levels, such as CX3CL1, CCL5, CXCL1, and CXCL10 pointing to T helper 1 (Th1) immune response [24]. CCL5, induced by IL1β and TNF‐α, suggests the excitation of Th1 [25, 26]. Several studies emphasize CCR7's significance in AA treatment, highlighting its therapeutic potential [27, 28, 29, 30]. CXCR6 plays a critical role in autoimmune diseases, triggering immune responses that result in tissue damage [31]. The CXCR6–CXCL16 axis governs immune cell functions, activating signaling pathways that increase cytokine secretion, including TNF‐α and IFN‐γ, aligning with our findings [32]. The study identified chemokine‐mediated signaling and cytokine‐cytokine receptor interaction as significant biological pathways in AA. Hair loss in AA is primarily caused by T cell activity, with CD4+T cells predominantly found around affected hair follicles [33, 34, 35]. Alterations in CD4+T cell subsets (Th1, Th2, Th17, and regulatory T cells) have been noted in AA. These cells, along with various cytokines and chemokines, contribute to the inflammatory response near hair follicles [36, 37]. Overexposed serum IL2, IFN‐γ, IL13, IL17A, and Th1 cytokines observed in AA patients [36, 37, 38]. The study identified key TFs regulating hub genes in AA, with ZNF683, FOXP3, and TBX21 influencing 9 out of 10 hub genes, while PRDM1and BATF regulated 6. ZNF683 has been linked to various skin conditions, including AA [39, 40, 41]. Hamed et al. found a decrease in FOXP3 CD39 Treg cells in their hair roots in individuals with AA [42]. Similarly, Speiser et al. found a substantial drop in the average mean prevalence of FOXP3 in individuals with AA [43], and TBX21 has an imperative role in AA pathogenesis [44]. BATF, identified as a significant immune system factor, is also involved in AA [45, 46, 47]. Variations in TF expression may influence these hub genes and act as risk factors in AA. Several TFs, including BATF, PRDM1, TCF7, NFATC2, SCML4, ZNF831, and TRAFD1, were predicted as key regulators in AA, corroborating findings from Roberts et al. who associated PRDM1 with autoimmune disorders [48]. In the present research, we effectively discovered 49 miRNAs that were found to be suitable for controlling the activity of seven particular genes related with AA. These miRNAs have been divided into several categories: 13 for CCR5, 13 for CD8A, 3 for CCL5, 4 for CXCR6, 4 for CXCL10, 10 for CXCL9, and 2 for CCL4. Papukashvili and his team's investigation on Androgenetic Alopecia (AGA) demonstrated that miRNAs in the Wnt/β‐catenin signaling pathway may promote hair development. MiRNAs that inhibit DKK1 can modulate the Wnt/β‐catenin signaling pathway, making them a promising area for research [49]. Sheng et al. discovered 36 significantly differently amplified miRNAs among individuals with severe active AA [50]. Furthermore, Wang and his colleagues underlined the importance of miRNAs in the setting of AA by studying variations in miRNAs engaging immune‐regulatory mechanisms [51]. They recommended miRNAs as fresh treatment possibilities for AA, which is backed by a number of previous investigations in the discipline [52, 53, 54]. Conclusion This study identified 10 TFs, 10 hub genes, and 49 miRNAs as potential aAA biomarkers. These genes influence AA progression via chemokine signaling, cytokine interactions, and related KEGG pathways, requiring further functional validation. Future Recommendation We recommend investigating genes' impact on AA progression through in vitro studies to validate predictive evaluations. This bioinformatics framework, combined with experimental research, offers a comprehensive approach for AA exploration. Study Limitations The relatively limited number of available samples may affect the generalizability and statistical robustness of the findings. Therefore, the conclusions drawn from the in silico analysis should be interpreted with caution until validated using larger and more diverse datasets. For understanding the role of these miRNAs, future research should therefore prioritize experimental validation—such as luciferase reporter assays, knockdown/overexpression studies, or analysis in patient‐derived samples—to confirm these regulatory interactions and clarify their contribution to AA. Author Contributions A.A., M.A., L.A., and S.B. collect the materials. A.A., M.A., S.B., L.A., and G.M. do the analysis and write the paper. A.A., M.A., S.B., L.A., and G.M. wrote the manuscript. A.A., M.A., S.B., L.A., and G.M. do the tables. Ethics Statement Institutional review board and Research Ethical Committee in accordance with the Helsinki Declaration guidelines. Conflicts of Interest The authors declare no conflicts of interest. Transparency Statement The lead author, Saleh S. Bahaj, affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained. Acknowledgments The authors have nothing to report. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. “All authors have read and approved the final version of the manuscript [Saleh Salem Bahaj] had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.” References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. “All authors have read and approved the final version of the manuscript [Saleh Salem Bahaj] had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.”