Searchlight: automated bulk RNA-seq exploration and visualisation using dynamically generated R scripts

Exploring RNA-Seq Data Analysis Through Visualization Techniques and Tools: A Systematic Review of Opportunities and Limitations for Clinical Applications

RNA sequencing (RNA-seq) has emerged as a prominent resource for transcriptomic analysis due to its ability to measure gene expression in a highly sensitive and accurate manner. With the increasing availability of RNA-seq data analysis from clinical studies and patient samples, the development of effective visualization tools for RNA-seq analysis has become increasingly important to help clinicians and biomedical researchers better understand the complex patterns of gene expression associated with health and disease. This review aims to outline the current state-of-the-art data visualization techniques and tools commonly used to frame clinical inferences from RNA-seq data and point out their benefits, applications, and limitations. A systematic review of English articles using PubMed, Scopus, Web of Science, and IEEE Xplore databases was performed. Search terms included "RNA-seq", "visualization", "plots", and "clinical". Only full-text studies reported between 2017 and 2024 were included for analysis. Following PRISMA guidelines, a total of 126 studies were identified, of which 33 studies met the inclusion criteria. We found that 18% of studies have visualization techniques and tools for circular RNA-seq data, 56% for single-cell RNA-seq data, 23% for bulk RNA-seq data, and 3% for long non-coding RNA-seq data. Overall, this review provides a comprehensive overview of the common visualization tools and their potential applications, which is a useful resource for researchers and clinicians interested in using RNA-seq data for various clinical purposes (e.g., diagnosis or prognosis).

Chronic inflammation degrades CD4 T cell immunity to prior vaccines in treated HIV infection

To date, our understanding of how HIV infection impacts vaccine-induced cellular immunity is limited. Here, we investigate inflammation, immune activation and antigen-specific T cell responses in HIV-uninfected and antiretroviral-treated HIV-infected people. Our findings highlight lower recall responses of antigen-specific CD4 T cells that correlate with high plasma cytokines levels, T cell hyperactivation and an altered composition of the T subsets enriched with more differentiated cells in the HIV-infected group. Transcriptomic analysis reveals that antigen-specific CD4 T cells of the HIV-infected group have a reduced expression of gene sets previously reported to correlate with vaccine-induced pathogen-specific protective immunity and further identifies a consistent impairment of the IFNα and IFNγ response pathways as mechanism for the functional loss of recall CD4 T cell responses in antiretroviral-treated people. Lastly, in vitro treatment with drugs that reduce inflammation results in higher memory CD4 T cell IFNγ responses. Together, our findings suggest that vaccine-induced cellular immunity may benefit from strategies to counteract inflammation in HIV infection.

CCR1 and CCR2 Coexpression on Monocytes Is Nonredundant and Delineates a Distinct Monocyte Subpopulation

The interactions between chemokines and their receptors, particularly in the context of inflammation, are complex, with individual receptors binding multiple ligands and individual ligands interacting with multiple receptors. In addition, there are numerous reports of simultaneous coexpression of multiple inflammatory chemokine receptors on individual inflammatory leukocyte subtypes. Overall, this has previously been interpreted as redundancy and proposed as a protective mechanism to ensure that the inflammatory response is robust. By contrast, we have hypothesized that the system is not redundant but exquisitely subtle. Our interests relate to the receptors CCR1, CCR2, CCR3, and CCR5, which, together, regulate nonneutrophilic myeloid cell recruitment to inflammatory sites. In this study, we demonstrate that although most murine monocytes exclusively express CCR2, there is a small subpopulation that is expanded during inflammation and coexpresses CCR1 and CCR2. Combinations of transcript and functional analysis demonstrate that this is not redundant expression and that coexpression of CCR1 and CCR2 marks a phenotypically distinct population of monocytes characterized by expression of genes otherwise typically associated with neutrophils. Single-cell RNA sequencing confirms this as a monodisperse population of atypical monocytes. This monocytic population has previously been described as having immunosuppressive activity. Overall, our data confirm combinatorial chemokine receptor expression by a subpopulation of monocytes but demonstrate that this is not redundant expression and marks a discrete monocytic population.

Stratifying macrophages based on their infectious burden identifies novel host targets for intervention during Crohn's disease associated adherent-invasive Escherichia coli infection

Bacterial infection is a dynamic process resulting in a heterogenous population of infected and uninfected cells. These cells respond differently based on their bacterial load and duration of infection. In the case of infection of macrophages with Crohn's disease (CD) associated adherent-invasive Escherichia coli (AIEC), understanding the drivers of pathogen success may allow targeting of cells where AIEC replicate to high levels. Here we show that stratifying immune cells based on their bacterial load identifies novel pathways and therapeutic targets not previously associated with AIEC when using a traditional homogeneous infected population approach. Using flow cytometry-based cell sorting we stratified cells into those with low or high intracellular pathogen loads, or those which were bystanders to infection. Immune cells transcriptomics revealed a diverse response to the varying levels of infection while pathway analysis identified novel intervention targets that were directly related to increasing intracellular AIEC numbers. Chemical inhibition of identified targets reduced AIEC intracellular replication or inhibited secretion of tumour necrosis factor alpha (TNFα), a key cytokine associated with AIEC infection. Our results have identified new avenues of intervention in AIEC infection that may also be applicable to CD through the repurposing of already available inhibitors. Additionally, they highlight the applicability of immune cell stratification post-infection as an effective approach for the study of microbial pathogens.

Co-inhibition of TGF-β and PD-L1 pathways in a metastatic colorectal cancer mouse model triggers interferon responses, innate cells and T cells, alongside metabolic changes and tumor resistance

Colorectal cancer (CRC) is the third most prevalent cancer worldwide with a high mortality rate (20-30%), especially due to metastasis to adjacent organs. Clinical responses to chemotherapy, radiation, targeted and immunotherapies are limited to a subset of patients making metastatic CRC (mCRC) difficult to treat. To understand the therapeutic modulation of immune response in mCRC, we have used a genetically engineered mouse model (GEMM), "KPN", which resembles the human 'CMS4'-like subtype. We show here that transforming growth factor (TGF-β1), secreted by KPN organoids, increases cancer cell proliferation, and inhibits splenocyte activation in vitro. TGF-β1 also inhibits activation of naive but not pre-activated T cells, suggesting differential effects on specific immune cells. In vivo, the inhibition of TGF-β inflames the KPN tumors, causing infiltration of T cells, monocytes and monocytic intermediates, while reducing neutrophils and epithelial cells. Co-inhibition of TGF-β and PD-L1 signaling further enhances cytotoxic CD8+T cells and upregulates innate immune response and interferon gene signatures. However, simultaneous upregulation of cancer-related metabolic genes correlated with limited control of tumor burden and/or progression despite combination treatment. Our study illustrates the importance of using GEMMs to predict better immunotherapies for mCRC.

Differences in transcriptional changes in psoriasis and psoriatic arthritis skin with immunoglobulin gene enrichment in psoriatic arthritis

OBJECTIVES

Approximately 20% of people with psoriasis develop PsA. Although genetic, clinical and environmental risk factors have been identified, it is not known why some people with psoriasis develop PsA. The skin disease is traditionally considered the same in both. This study compares transcriptional changes in psoriasis and PsA skin for the first time.

METHODS

Skin biopsies were collected from healthy controls (HC), and uninvolved and lesional skin from patients with PsA. Bulk tissue sequencing was performed and analysed using the pipeline Searchlight 2.0. Transcriptional changes in PsA skin were compared with existing sequencing data from participants with psoriasis without PsA (GSE121212). Psoriasis and PsA datasets could not be directly compared as different analysis methods were used. Data from participants with PsA in the GSE121212 dataset were used for validation.

RESULTS

Skin samples from 9 participants with PsA and 9 HC were sequenced, analysed and compared with available transcriptomic data for 16 participants with psoriasis compared with 16 HC. Uninvolved skin in psoriasis shared transcriptional changes with lesional skin in psoriasis, but uninvolved skin in PsA did not. Most transcriptional changes in psoriasis and PsA lesional skin were shared, but immunoglobulin genes were upregulated in PsA lesional skin specifically. The transcription factor POU2F1, which regulates immunoglobulin gene expression, was enriched in PsA lesional skin. This was confirmed in the validation cohort.

CONCLUSIONS

Immunoglobulin genes are upregulated in PsA but not in psoriasis skin lesions. This may have implications for the spread from the cutaneous compartment to other tissues.

Differential regulation of IL-17A and IL-17F via STAT5 contributes to psoriatic disease

BACKGROUND

IL-17A plays a pivotal pathogenic role in several immune-mediated inflammatory diseases. Despite sharing 50% sequence homology with IL-17A, the role of IL-17F remains less clear. Clinical findings suggest that dual inhibition of IL-17A and IL-17F in psoriatic disease is more efficacious than IL-17A inhibition alone, positing a pathogenic role for IL-17F.

OBJECTIVE

We characterized the regulation of IL-17A and IL-17F in psoriatic disease.

METHODS

Using both in vitro systems and lesional skin tissue from patients, we interrogated the chromosomal, transcriptional, and protein expression landscape of IL-17A+ and IL-17F+ TH17 cells. Alongside established assays such as single-cell RNA sequencing, we developed a novel cytokine-capture technique that was combined with chromatin immunoprecipitation sequencing and RNA sequencing.

RESULTS

We confirm a preferential elevation of IL-17F over IL-17A in psoriatic disease and show that expression of each isoform predominantly occurs in distinct cell populations. The expression of both IL-17A and IL-17F exhibited a high degree of plasticity, with the balance between the 2 isoforms influenced by proinflammatory signaling and by anti-inflammatory drugs such as methylprednisolone. This plasticity was reflected in a broad H3K4me3 region at the IL17A-F locus, while opposing effects of STAT5/IL-2 signaling were observed for each of the 2 genes. Functionally, higher IL17F expression was linked to greater cell proliferation.

CONCLUSION

There are key differences in the regulation of IL-17A and IL-17F in psoriatic disease, leading to distinct inflammatory cell populations. As such, we propose that both IL-17A and IL-17F neutralization may be required to maximally inhibit IL-17-driven pathology.

Cutaneous lesions in psoriatic arthritis are enriched in chemokine transcriptomic pathways

OBJECTIVES

Skin from people with psoriasis has been extensively studied and is assumed to be identical to skin from those with psoriatic arthritis (PsA). Chemokines and the CC chemokine scavenger receptor ACKR2 are upregulated in uninvolved psoriasis. ACKR2 has been proposed as a regulator of cutaneous inflammation in psoriasis. The aim of this study was to compare the transcriptome of PsA skin to healthy control (HC) skin and evaluate ACKR2 expression in PsA skin.

METHODS

Full-thickness skin biopsies from HC, lesional and uninvolved skin from participants with PsA were sequenced on NovaSeq 6000. Findings were validated using qPCR and RNAscope.

RESULTS

Nine HC and nine paired PsA skin samples were sequenced. PsA uninvolved skin was transcriptionally similar to HC skin, and lesional PsA skin was enriched in epidermal and inflammatory genes. Lesional PsA skin was enriched in chemokine-mediated signalling pathways, but uninvolved skin was not. ACKR2 was upregulated in lesional PsA skin but had unchanged expression in uninvolved compared with HC skin. The expression of ACKR2 was confirmed by qPCR, and RNAscope demonstrated strong expression of ACKR2 in the suprabasal layer of the epidermis in PsA lesions.

CONCLUSION

Chemokines and their receptors are upregulated in lesional PsA skin but relatively unchanged in uninvolved PsA skin. In contrast to previous psoriasis studies, ACKR2 was not upregulated in uninvolved PsA skin. Further understanding of the chemokine system in PsA may help to explain why inflammation spreads from the skin to the joints in some people with psoriasis.

bulkAnalyseR: an accessible, interactive pipeline for analysing and sharing bulk multi-modal sequencing data

Bulk sequencing experiments (single- and multi-omics) are essential for exploring wide-ranging biological questions. To facilitate interactive, exploratory tasks, coupled with the sharing of easily accessible information, we present bulkAnalyseR, a package integrating state-of-the-art approaches using an expression matrix as the starting point (pre-processing functions are available as part of the package). Static summary images are replaced with interactive panels illustrating quality-checking, differential expression analysis (with noise detection) and biological interpretation (enrichment analyses, identification of expression patterns, followed by inference and comparison of regulatory interactions). bulkAnalyseR can handle different modalities, facilitating robust integration and comparison of cis-, trans- and customised regulatory networks.

PredPromoter-MF(2L): A Novel Approach of Promoter Prediction Based on Multi-source Feature Fusion and Deep Forest

Promoters short DNA sequences play vital roles in initiating gene transcription. However, it remains a challenge to identify promoters using conventional experiment techniques in a high-throughput manner. To this end, several computational predictors based on machine learning models have been developed, while their performance is unsatisfactory. In this study, we proposed a novel two-layer predictor, called PredPromoter-MF(2L), based on multi-source feature fusion and ensemble learning. PredPromoter-MF(2L) was developed based on various deep features learned by a pre-trained deep learning network model and sequence-derived features. Feature selection based on XGBoost was applied to reduce fused features dimensions, and a cascade deep forest model was trained on the selected feature subset for promoter prediction. The results both fivefold cross-validation and independent test demonstrated that PredPromoter-MF(2L) outperformed state-of-the-art methods.

Dissecting the molecular control of immune cell accumulation in the inflamed joint

Mechanisms governing entry and exit of immune cells into and out of inflamed joints remain poorly understood. We sought herein to identify the key molecular pathways regulating such migration. Using murine models of inflammation in conjunction with mice expressing a photoconvertible fluorescent protein, we characterized the migration of cells from joints to draining lymph nodes and performed RNA-Seq analysis on isolated cells, identifying genes associated with migration and retention. We further refined the gene list to those specific for joint inflammation. RNA-Seq data revealed pathways and genes previously highlighted as characteristic of rheumatoid arthritis in patient studies, validating the methodology. Focusing on pathways associated with cell migration, adhesion, and movement, we identified genes involved in the retention of immune cells in the inflamed joint, namely junctional adhesion molecule A (JAM-A), and identified a role for such molecules in T cell differentiation in vivo. Thus, using a combination of cell-tracking approaches and murine models of inflammatory arthritis, we identified genes, pathways, and anatomically specific tissue signatures regulating cell migration in a variety of inflamed sites. This skin- and joint-specific data set will be an invaluable resource for the identification of therapeutic targets for arthritis and other inflammatory disorders.