Researchers in the North East of the United Kingdom have made a significant advancement in understanding a common yet often misunderstood chronic condition. By harnessing artificial intelligence (AI), experts from Newcastle University and collaborators have identified key genetic patterns that could revolutionise treatment approaches. This development not only addresses the skin manifestations but also the serious cardiovascular risks associated with the disease, paving the way for personalised care strategies.

Psoriasis: A Chronic Inflammatory Condition with Far-Reaching Health Impacts

Psoriasis, a chronic autoimmune disease characterised by rapid skin cell growth leading to thick, scaly plaques, affects approximately 2% of the UK population, equating to over 1.3 million people. While primarily known for its visible skin symptoms—itchy, painful red patches that can cover large areas of the body—the condition's dangers extend far beyond dermatological concerns. Systemic inflammation driven by psoriasis significantly elevates the risk of comorbidities, including type 2 diabetes, psoriatic arthritis, and crucially, cardiovascular diseases such as heart attacks and strokes.

In the North East of England, where socioeconomic challenges contribute to higher rates of lifestyle-related illnesses, the prevalence of psoriasis and its complications is particularly acute. Public Health England data indicates that coronary heart disease mortality rates in the region are among the highest in the country, with inflammation from conditions like psoriasis exacerbating this burden. Patients often face stigma, with historical accounts of individuals being excluded from public spaces like swimming pools due to misconceptions about the contagious nature of their lesions. This social isolation compounds the physical toll, underscoring the need for innovative research from regional universities.

The inflammatory process in psoriasis involves overactive T-cells triggering cytokine release, such as tumour necrosis factor-alpha (TNF-α) and interleukin-17 (IL-17), which not only accelerate keratinocyte proliferation but also promote atherosclerosis—the buildup of plaques in arteries that can lead to myocardial infarction (heart attack). Step-by-step, this begins with genetic predispositions interacting with environmental triggers like stress, infections, or obesity, culminating in endothelial dysfunction and heightened clotting risk.

The PSORT Consortium: Pioneering Multi-Modal Research

The PSORT (Psoriasis Stratification to Optimise Relevant Therapy) Consortium represents a collaborative effort between Newcastle University, Queen Mary University of London, King's College London, and industry partners. Funded primarily by the Medical Research Council (MRC), this initiative has produced one of the largest datasets on psoriasis, comprising over 700 samples from blood, lesional skin (affected areas), and non-lesional skin (unaffected areas) of patients initiating biologic therapies.

Led by Professor Nick Reynolds, Dermatology Professor and Director of Diagnostics at Newcastle University's Faculty of Medical Sciences, and Professor Mike Barnes from Queen Mary, the team integrated clinical outcomes with transcriptomic data— the study of all RNA molecules to capture gene activity snapshots. This comprehensive approach allowed for the identification of molecular endotypes, or subtypes, within psoriasis, moving beyond one-size-fits-all treatments.

Newcastle Hospitals NHS Foundation Trust, where Prof Reynolds serves as an honorary consultant dermatologist, provided real-world patient data, highlighting the synergy between academia and the NHS. Such partnerships are vital in UK higher education, fostering translational research that directly benefits patients while offering training opportunities for PhD students and postdocs in bioinformatics and genetics.

• Dataset scale: >700 multi-omics samples
• Patient cohort: Those starting biologic drugs like adalimumab
• Collaborators: NIHR Newcastle Biomedical Research Centre, Psoriasis Association

Harnessing AI and Machine Learning for Genetic Insights

At the core of this breakthrough is the application of machine learning (ML), a subset of AI that uses algorithms to detect patterns in vast datasets without explicit programming. The researchers employed unsupervised clustering techniques, such as principal component analysis (PCA) and t-distributed stochastic neighbour embedding (t-SNE), to group gene expression profiles. Supervised models, including random forests and neural networks, then correlated these with clinical severity scores like the Psoriasis Area and Severity Index (PASI).

Step-by-step process:

1. Data preprocessing: Normalisation of RNA-seq data to account for batch effects and sequencing depth.
2. Feature selection: Identification of differentially expressed genes (DEGs) using DESeq2.
3. ML modelling: Training models on lesional vs. non-lesional data to uncover signatures.
4. Validation: Cross-validation and external cohorts to ensure robustness.
5. Interpretation: Pathway enrichment analysis revealing neutrophil degranulation and metabolic pathways.

This AI-driven methodology not only pinpointed novel biomarkers but also made the data publicly accessible via an interactive web portal, empowering global researchers. For aspiring data scientists in higher education, this exemplifies how computational biology intersects with clinical medicine.

Explore research jobs in AI and genomics at leading UK universities.

Photo by Google DeepMind on Unsplash

Breakthrough Findings: The Nine-Gene Biomarker

The study's flagship discovery is a nine-gene biomarker strongly associated with psoriasis severity. These genes, involved in immune regulation and skin barrier function, showed consistent upregulation in severe cases, independent of treatment response. Specific HLA variants—HLA-DQA1*01 and HLA-DRB1*15—were linked to baseline disease harshness, providing heritable risk markers.

Another key insight is a 14-gene signature tied to body mass index (BMI) in non-lesional skin, which also influenced lesional severity. This underscores obesity's role as a modifiable factor amplifying inflammation via adipokines like leptin. Post-adalimumab treatment, a blood-based signal from white blood cells emerged, indicating targeted immune modulation.

These findings, detailed in Communications Medicine, offer concrete examples: patients with the biomarker profile might prioritise IL-17 inhibitors over TNF blockers, optimising outcomes. Read the full study.

From Stigma to Solutions: Addressing Psoriasis in the North East

In the North East, psoriasis prevalence mirrors national averages, but access to biologics lags due to funding pressures. The Newcastle-led research addresses this by proposing stratified care, potentially reducing trial-and-error in therapy selection. Current pathways involve topical treatments, phototherapy, then systemic drugs—a process taking months.

Stakeholder perspectives vary: The Psoriasis Association praises the hope for tailored therapies, while patients report relief from stigma through education. Prof Reynolds emphasised: "We waste a lot of time... If we can use this technology to change that, it'd make a lot of difference."

• Benefits of personalised care: Faster remission, fewer side effects, cost savings for NHS
• Risks of delay: Persistent inflammation heightens CVD events by 50% in severe cases

This aligns with UK higher education's push for impact-driven research, as seen in REF assessments.

Newcastle University's Role in Medical Innovation

Newcastle University, ranked top 20 in the UK for medicine, hosts world-class facilities like the Newcastle NIHR Biomedical Research Centre. Prof Reynolds' team exemplifies this, blending dermatology with genomics. Opportunities abound for faculty positions and postdoc roles in translational medicine.

Collaborations with Queen Mary enhance multi-omics expertise, training the next generation in AI ethics and data privacy—crucial amid GDPR concerns.

Implications for Cardiovascular Health and Personalised Medicine

Psoriasis doubles heart attack risk via shared inflammatory pathways like IL-6 and C-reactive protein elevation. The genetic insights could inform polygenic risk scores for CVD in psoriasis patients, integrating with tools like QRISK3.

Personalised medicine—matching drugs to genetic profiles—mirrors oncology successes with PARP inhibitors. Future trials may validate these biomarkers prospectively.

For career advice in this field, visit higher ed career advice.

Photo by Google DeepMind on Unsplash

Challenges, Ethical Considerations, and Future Outlook

Challenges include data biases in AI training and equitable access in deprived areas. Ethically, genetic testing raises privacy issues, addressed via federated learning.

Outlook: Expanded PSORT cohorts, AI integration into NHS apps, and pan-European trials. North East universities position UK as a leader in precision health.

Opportunities in Higher Education Research

This study highlights booming demand for experts in bioinformatics. Lecturer jobs and professor jobs in genetics are plentiful. Rate professors via Rate My Professor.

In conclusion, this AI-driven discovery heralds a new era. Explore higher ed jobs, career advice, and rate your professors to join the revolution.