Advancing protein design is crucial for breakthroughs in medicine and biotechnology. Traditional approaches for protein sequence representation often rely solely on the 20 canonical amino acids, limiting the representation of non-canonical amino acids and residues that undergo post-translational modifications. This work explores discrete diffusion models for generating novel protein sequences using the all-atom chemical representation SELFIES. By encoding the atomic composition of each amino acid in the protein, this approach expands the design possibilities beyond standard sequence representations. Using a modified ByteNet architecture within the discrete diffusion D3PM framework, we evaluate the impact of this all-atom representation on protein quality, diversity, and novelty, compared to conventional amino acid-based models. To this end, we develop a comprehensive assessment pipeline to determine whether generated SELFIES sequences translate into valid proteins containing both canonical and non-canonical amino acids. Additionally, we examine the influence of two noise schedules within the diffusion process—uniform (random replacement of tokens) and absorbing (progressive masking)—on generation performance. While models trained on the all-atom representation struggle to consistently generate fully valid proteins, the successfully generated proteins show improved novelty and diversity compared to their amino acid-based model counterparts. Furthermore, the all-atom representation achieves structural foldability results comparable to those of amino acid-based models. Lastly, our results highlight the absorbing noise schedule as the most effective for both representations. Data and code are available at https://github.com/Intelligent-molecular-systems/All-Atom-Protein-Sequence-Generation.

Many molecular aging biomarkers have been developed to capture heterogeneity in individual aging rates. Yet, systematic comparison of the modeling choices underlying these biomarkers has been limited. In this study, we trained aging biomarkers on the Rockwood frailty index (FI) and all-cause mortality using UK Biobank Olink proteomics and metabolomics (¹H-NMR) data (n = 40,696). We systematically established the impact of model choice, target outcome, and molecular data source on several age-related outcomes. From this, we developed two aging biomarkers, ProteinFrailty (ProtFI) and ProteinMortality (ProtMort), which are both ElasticNet models that use a minimal set of proteins to predict FI and mortality, respectively. In particular, ProtFI outperformed established aging biomarkers in relation to diverse outcomes, including incident cardiovascular disease, handgrip strength, and self-rated health, both in internal validation and two Dutch external cohorts (n = 995, n = 500). Our findings show that an efficient frailty-trained proteomic biomarker robustly predicts age-related decline.

BACKGROUND: Alternative splicing contributes to molecular diversity across brain cell types. RNA-binding proteins (RBPs) regulate splicing, but the genome-wide mechanisms underlying cell-type-specific splicing remain poorly understood. RESULTS: Here, we want to unravel cell-type-specific splicing mechanisms by using RBP binding sites and/or the genomic sequence to predict exon inclusion in neurons and glia as measured by long-read single-cell data in the human hippocampus and frontal cortex. We found that exon inclusion of variable exons is harder to predict in neurons compared to glia in both brain regions. Comparing neurons and glia, the position of RBP binding sites in alternatively spliced exons in neurons differ more from non-variable exons indicating distinct splicing mechanisms. Model interpretation pinpointed RBPs, including QKI, potentially regulating alternative splicing between neurons and glia. Finally, we accurately predict and prioritize the effect of splicing QTLs. CONCLUSIONS: Our results indicate that the splicing mechanisms in variable exons in neurons diverged more from the standard mechanisms. Splicing in neurons might be less sequence-dependent and influenced more by, for instance, chromatin accessibility or methylation. Taken together, these results highlight new insights into the mechanisms regulating cell-type-specific alternative splicing in the brain.

Rheumatic diseases are chronic, immune-mediated conditions characterized by significant heterogeneity in presentation and disease course. However, current clinical approaches often rely on snapshot-based assessments that fail to capture the complex longitudinal evolution of these conditions. To address these limitations and support the implementation of precision medicine, we present the design for the Rheumatic Digital Twin, a novel, modular conceptual framework intended to integrate heterogeneous multimodal data, ranging from electronic health records and clinical notes to imaging and omics, into a dynamic, computational representation of the patient journey. Our theoretical architecture addresses challenges related to data silos and variable availability of data modalities through a multistage approach that envisions the use of domain-specific foundation models to independently process distinct data modalities. To effectively model the temporal progression inherent in chronic diseases, the proposed design utilizes Transformer architectures, leveraging self-attention mechanisms to treat patient events, such as lab results or medication changes, as sequential data tokens. We describe how these unimodal representations would subsequently be fused via joint embedding techniques to construct a shared, multimodal representational space. Envisioned to function analogously to a recommender system, the Rheumatic Digital Twin framework is modeled to map patients into a latent space where proximity reflects clinical and biological similarity. By identifying “nearest neighbors,” historical patients with comparable trajectories, the system aims to enable in silico cohorting, theoretically allowing clinicians to forecast key clinical events, predict treatment responses, and identify likely disease courses based on the outcomes of similar peers.

Background: Nutritional weight-loss interventions are known to reduce bone mineral density (BMD), which can be prevented by adding (resistance) exercise training. However, this combined effect is not well studied in non-obese adults. In addition, the association between biomarkers and metabolite-based composite health markers with changes in BMD in such an intervention has not been studied as thoroughly. Objective: The aims of the current study were to investigate the effect of a combined nutritional and activity lifestyle intervention on lumbar spine and total body BMD in healthy middle-aged to older adults, and to relate these effects to a selection of immune-metabolic biomarkers, muscle mass and fat mass measurements, and two composite metabolite-based health scores. Methods: In this ancillary study of the single-arm Growing Old TOgether (GOTO) trial (trial registration number GOTNL3301 [https://onderzoekmetmensen.nl/nl/trial/27183], NL-OMON27183), 134 participants (mean age 62.9 years, 49% female) undertook a 13-week lifestyle modification, incorporating 12.5% caloric restriction and 12.5% increase in physical activity. The impact on lumbar spine and total body BMD was evaluated using dual-energy X-ray absorptiometry (DEXA). The intervention effect on BMD was related to changes in immune-metabolic biomarkers and two metabolite-based immune-metabolic health scores. Results: The trial significantly reduced bodyweight with 3.3 and 3.4 kg, consisting of 1.4 and 1.1 kg lean mass, in males (fdr < 0.001) and females (fdr < 0.001), respectively. Lean mass reduced by 1.4 kg in males (fdr < 0.001) and 1.1 kg in females (fdr < 0.001), whereas total body fat% reduced significantly with −1.5% (fdr < 0.001) in males and −1.5% (fdr < 0.001) in females. In males, lumbar spine BMD increased with 3.0% (fdr < 0.001) and total body BMD with 0.7% (fdr = 0.002). In females, the lumbar spine BMD had a trend in the upwards direction (1.2%, fdr = 0.09) and the total body BMD remained stable (0.4%, fdr = 0.07). In males, the increase in lumbar spine BMD was significantly associated with decreased weight (fdr = 0.001) and with decreased body and trunk fat% (fdr = 0.001, fdr = 0.001) and improved immune-metabolic health (fdr = 0.02). Males with higher BMD but a poor metabolite-based health score at baseline had a stronger increase in lumbar spine BMD (fdr = 0.03). Conclusions: A combined nutritional and activity lifestyle intervention significantly improved BMD of males with good bone health at baseline while at the same time improving metabolic health. Nutritional weight-loss interventions may not harm BMD when combined with exercise.

Transcriptomic differences between individuals and sexes are well-documented across tissues, affecting cell-type identity. Single-cell atlases often have skewed sex ratios or limited donor diversity, potentially leading to sex- or donor-biased annotations using automatic classification methods. This might cause models to exacerbate existing biases from their training data. We investigated this by varying the sex ratio in a training set and assessing a cell-type classifier’s performance on fixed single-sex test sets. To separate sex and donor effects, we ran the experiment with and without donor information. We found that differences between donors negatively impact classification, as evidenced by poorer performance on unseen donors. However, this is not primarily due to sex bias; the model classified male and female cells similarly well, even when trained on highly sex-skewed data. We also found that large sex-based abundance differences between cell types can confound performance interpretation, creating apparent sex-biased patterns. Our findings suggest that atlas creators and classifier developers must carefully consider donor-specific biases in scRNA-seq data.

Federated learning is an upcoming machine learning paradigm which allows data from multiple sources to be used for training of classifiers without the data leaving the source it originally resides. This can be highly valuable for use cases such as medical research, where gathering data at a central location can be quite complicated due to privacy and legal concerns of the data. In such cases, federated learning has the potential to vastly speed up the research cycle. Although federated and central learning have been compared from a theoretical perspective, an extensive experimental comparison of performances and learning behavior still lacks. We have performed a comprehensive experimental comparison between federated and centralized learning. We evaluated various classifiers on various datasets exploring influences of different sample distributions as well as different class distributions across the clients. The results show similar performances under a wide variety of settings between the federated and central learning strategies. Federated learning is able to deal with various imbalances in the data distributions. It is sensitive to batch effects between different datasets when they coincide with location, similar to central learning, but this setting might go unobserved more easily. Federated learning seems to be robust to various challenges such as skewed data distributions, high data dimensionality, multiclass problems, and complex models. Taken together, the insights from our comparison gives much promise for applying federated learning as an alternative to sharing data. Code for reproducing the results in this work can be found at: https://github.com/swiergarst/FLComparison.

Background: Despite several randomised controlled trials (RCTs) on the use of adjuvant treatment with corticosteroids in patients with community-acquired pneumonia (CAP), the effect of this intervention on mortality remains controversial. We aimed to evaluate heterogeneity of treatment effect (HTE) of adjuvant treatment with corticosteroids on 30-day mortality in patients with CAP. Methods: In this individual patient data meta-analysis, we included RCTs published before July 1, 2024, comparing adjuvant treatment with corticosteroids versus placebo in patients hospitalised with CAP. The primary endpoint was 30-day all-cause mortality, collected across all trials, and analyses followed the intention-to-treat principle. We analysed HTE using risk and effect modelling. For risk modelling, patients were classified as having less severe or severe CAP based on the pneumonia severity index (PSI), comparing PSI class I–III versus class IV–V. For effect modelling, we trained a corticosteroid-effect model on six trials and externally validated it using data from two trials, received after model preregistration. This model classified patients into two groups: no predicted benefit and predicted benefit from adjuvant treatment with corticosteroids. The literature search was registered on PROSPERO, CRD42022380746. Findings: We included eight RCTs with 3224 patients. Across all eight trials, 246 (7·6%) patients died within 30 days (106 [6·6%] of 1618 in the corticosteroid group vs 140 [8·7%] of 1606 in the placebo group; odds ratio [OR] 0·72 [95% CI 0·56–0·94], p=0·017). The corticosteroid-effect model, which selected C-reactive protein (CRP), showed significant HTE during external validation in the two most recent trials. In these trials, 154 (11·4%) of 1355 patients died within 30 days (88 [13·1%] of 671 in the placebo group vs 66 [9·6%] of 684 in the corticosteroid group; OR 0·71 [95% CI 0·50–0·99], p=0·044). Among patients predicted to have no benefit (CRP ≤204 mg/L, n=725), no significant effect was observed (OR 0·98 [95% CI 0·63–1·50]), whereas for those with predicted benefit (CRP >204 mg/L, n=630), 39 (13·0%) of 301 patients died in the placebo group compared with 20 (6·1%) of 329 in the corticosteroid group (0·43 [0·25–0·76], p_interaction=0·026). No significant HTE was found between less severe CAP (PSI class I–III, n=229) and severe CAP (PSI class IV–V, n=1126). Corticosteroid therapy significantly increased hyperglycaemia risk (44 [12·8%] of 344 in the placebo group vs 84 [24·8%] of 339 in the corticosteroid group; OR 2·50 [95% CI 1·63–3·83], p<0·0001) and hospital re-admission risk (30 [3·7%] of 814 in the placebo group vs 57 [7·0%] of 819 in the corticosteroid group; 1·95 [1·24–3·07], p=0·0038). Interpretation: Overall, adjuvant therapy with corticosteroids significantly reduces 30-day mortality in patients hospitalised with CAP. The treatment effect varied significantly among subgroups based on CRP concentrations, with a substantial mortality reduction observed only in patients with high baseline CRP. Funding: None.

Plants respond to stresses like drought and heat through complex gene regulatory networks (GRNs). To improve resilience, understanding these is crucial, but large-scale GRNs (>100 genes) are difficult to model using ordinary differential equations (ODEs) due to the high number of parameters that have to be estimated. Here we solve this problem by introducing BADDADAN, which uses machine learning to identify gene modules—groups of co-expressed and/or co-regulated genes—and constructs an ODE model that predicts gene module dynamics under stress. By integrating time-series gene expression data with prior co-expression data it finds modules that are both coherent and interpretable. We demonstrate BADDADAN on heat and drought datasets of A. thaliana, modelling over 1,000 genes, recovering known mechanistic insights, and proposing new hypotheses. By combining machine learning with mechanistic modelling, BADDADAN deepens our understanding of stress-related GRNs in plants and potentially other organisms.

Background: Fractional exhaled nitric oxide (FeNO) is a noninvasive method to determine the degree of airway inflammation. Handheld devices such as the Vivatmo Me are used for home monitoring. Differences were found between the Vivatmo Me and standard measurements with the NIOX VERO. Therefore, we aimed to determine the accuracy of the Vivatmo Me for FeNO measurements. Methods: Adult patients with an appointment for FeNO-measurement according to regular care, were invited to perform the FeNO measurement with both devices. From these measurements the FeNO values were compared, and the device user-friendliness was determined. Results: One hundred and sixty-four patients were included. The number of attempts needed for a successful measurement and the failure rate were higher with the Vivatmo Me. Although the measurements were highly correlated, a significant difference (p < 0.001) was found between FeNO values measured with both devices. From the Vivatmo measurements, 32% did not fall within the claimed accuracy ranges. A linear correction on the FeNO values reduced this number. Conclusion: Our findings indicate that the Vivatmo Me does not comply with the claimed accuracy of clinical FeNO measurements and the measurement is challenging to perform. By applying the proposed correction, the comparative validity of the FeNO measurement improves and therefore its clinical usefulness.

Background and ObjectivesIdentifying genetic causes of dementia in patients visiting memory clinics is important for patient care and family planning. Traditional clinical selection criteria for genetic testing may miss carriers of pathogenic variants in dementia-related genes. This study aimed identify how many carriers we are missing and to optimize criteria for selecting patients for genetic counseling in memory clinics.MethodsIn this clinical cohort study, we retrospectively genetically tested patients during 2.5 years (2010-2012) visiting the Alzheimer Center Amsterdam, a specialized memory clinic. Genetic tests consisted of a 54-gene dementia panel, focusing on Class IV/V variants per American College of Medical Genetics and Genomics guidelines, including APP duplications and the C9ORF72 repeat expansion. We determined the prevalence of pathogenic variants and propose new eligibility criteria for genetic testing in memory clinics. The eligibility criteria were prospectively applied for 1 year (2021-2022), and results were compared with the retrospective cohort.ResultsGenetic tests were retrospectively performed in in 1,022 of 1,138 patients (90%) who consecutively visited the memory clinic. Among these, 1,022 patients analyzed (mean age 62.1 ± 8.9 years; 40.4% were female), 34 pathogenic variant carriers were identified (3.3%), with 24 being symptomatic. Previous clinical criteria would have identified only 15 carriers (44% of all carriers, 65% of symptomatic carriers). The proposed criteria increased identification to 22 carriers (62.5% of all carriers, 91% of symptomatic carriers). In the prospective cohort, 148 (28.7%) of 515 patients were eligible for testing under the new criteria. Of the 90 eligible patients who consented to testing, 13 pathogenic carriers were identified, representing a 73% increase compared with the previous criteria.DiscussionWe found that patients who visit a memory clinic and carry a pathogenic genetic variant are often not eligible for genetic testing. The proposed new criteria improve the identification of patients with a genetic cause for their cognitive complaints. In systems without practical or financial barriers to genetic testing, the new criteria can enhance personalized care. In other countries where the health care systems differs and in other genetic ancestry groups, the performance of the criteria may be different.