Search Results (24)

Background/Objectives: The Janus kinase-signal transducer and activator of transcription (JAK-STAT) signaling pathway is a critical mediator of immune regulation, inflammation, and cancer progression. Although implicated in colorectal cancer (CRC) pathogenesis, its molecular heterogeneity and clinical significance remain insufficiently characterized—particularly within early-onset CRC (EOCRC) and across diverse treatment and demographic contexts. We present AI-HOPE-JAK-STAT, a novel conversational artificial intelligence platform built to enable the real-time, natural language-driven exploration of JAK/STAT pathway alterations in CRC. The platform integrates clinical, genomic, and treatment data to support dynamic, hypothesis-generating analyses for precision oncology. Methods: AI-HOPE-JAK-STAT combines large language models (LLMs), a natural language-to-code engine, and harmonized public CRC datasets from cBioPortal. Users define analytical queries in plain English, which are translated into executable code for cohort selection, survival analysis, odds ratio testing, and mutation profiling. To validate the platform, we replicated known associations involving JAK1, JAK3, and STAT3 mutations. Additional exploratory analyses examined age, treatment exposure, tumor stage, and anatomical site. Results: The platform recapitulated established trends, including improved survival among EOCRC patients with JAK/STAT pathway alterations. In FOLFOX-treated CRC cohorts, JAK/STAT-altered tumors were associated with significantly enhanced overall survival (p < 0.0001). Stratification by age revealed survival advantages in younger (age < 50) patients with JAK/STAT mutations (p = 0.0379). STAT5B mutations were enriched in colon adenocarcinoma and correlated with significantly more favorable trends (p = 0.0000). Conversely, JAK1 mutations in microsatellite-stable tumors did not affect survival, emphasizing the value of molecular context. Finally, JAK3-mutated tumors diagnosed at Stage I–III showed superior survival compared to Stage IV cases (p = 0.00001), reinforcing stage as a dominant clinical determinant. Conclusions: AI-HOPE-JAK-STAT establishes a new standard for pathway-level interrogation in CRC by empowering users to generate and test clinically meaningful hypotheses without coding expertise. This system enhances access to precision oncology analyses and supports the scalable, real-time discovery of survival trends, mutational associations, and treatment-response patterns across stratified patient cohorts. Full article

This study analyzes publicly available RNA-seq data to comprehensively include the complex heterogeneity of prostate cancer (PCa) etiology. It combines prostate and prostate cancer (PCa) cell lines, representing primary PCa cells, Gleason scores, ages, and PCa of different racial origins. Additionally, some cell lines were exposed to endocrine-disrupting chemicals (EDCs). The research aims to identify hub genes and transcription factors (TFs) of the prostate carcinogenesis pathway as molecular targets for clinical investigations to elucidate EDC-induced aggressiveness and to develop potential biomarkers for their exposure risk assessments. PCa cells rely on androgen receptor (AR)-mediated signaling to survive, develop, and function. Fifteen various RNA-seq datasets were normalized for distribution, and the significance (p-value < 0.05) threshold of differentially expressed genes (DEGs) was set based on |log2FC| ≥ 2 change. Through integrated bioinformatics, we applied cBioPortal, UCSC-Xena, TIMER2.0, and TRRUST platforms, among others, to associate hub genes and their TFs based on their biologically meaningful roles in aggressive prostate carcinogenesis. Among all RNA-Seq datasets, we found 75 overlapping DEGs, with BUB1B (32%) and CCNB1 (29%) genes exhibiting the highest degree of mutation, amplification, and deletion. EDC-associated CCNB1, BUB1B, and CCNA2 in PCa cells exposed to EDCs were consistently shown to be associated with high Gleason scores (≥4 + 3) and in the >60 age group of patients. Selected TFs (E2F4, MYC, and YBX1) were also significantly associated with DEGs (NCAPG, MKI67, CCNA2, CCNB1, CDK1, CCNB2, AURKA, UBE2C, BUB1B) and influenced the overall survival (p-value < 0.05) of PCa cases. This is one of the first comprehensive studies combining 15 publicly available RNA-seq datasets to demonstrate the association of EDC-associated hub genes and their TFs aligning with the aggressive carcinogenic pathways in the higher age group (>60 years) of patients. The findings highlight the potential of these hub genes as candidates for further studies to develop molecular biomarkers for assessing the EDC-related PCa risk, diagnosing PCa aggressiveness, and identifying therapeutic targets. Full article

Prostate cancer (PCa) is the second most common cancer and the second leading cause of cancer-related mortality among men. Gene expression analysis has been crucial in understanding tumor biology and providing disease progression markers. Cell surface glycoproteins and those in the extracellular matrix play significant roles in the PCa microenvironment by promoting migration, invasion, and metastasis. The molecular and histopathological heterogeneity of prostate tumors necessitates a new marker discovery to better stratify patients at risk for poor prognosis. In this study, our objectives were to investigate and characterize the localization and expression of SLIT/ROBO in PCa samples from transgenic mice and human tumor samples, aiming to identify novel prognostic markers and potential therapeutic targets. We conducted histopathological, immunohistochemical, and bioinformatics analyses on prostate tumors from two knockout mice models (Pb-Cre4/Pten^f/f and Pb-Cre4/Trp53^f/f;Rb1^f/f) and human prostate tumors. Transcriptomic analyses revealed special changes in the expression of genes related to the SLIT/ROBO neural signaling pathway. We further characterized the gene and protein expression of the SLIT/ROBO pathway in knockout animal samples, and protein expression in the PCa samples of patients with different Gleason scores. Public datasets with clinical data from patients (The Human Protein Atlas, cBioPortal, SurvExpress and CamcAPP) were used to validate the gene and protein expression of SLIT1, SLIT2, ROBO1, and ROBO4, correlating these alterations with the prognosis of subgroups of patients. Our findings highlight potential biomarkers of the SLIT/ROBO pathway with prognostic and predictive value, as well as promising therapeutic targets for PCa. Full article

The development of digital instruments for mental health monitoring using biosensor data from wearable devices can enable remote, longitudinal, and objective quantitative benchmarks. To survey developments and trends in this field, we conducted a systematic review of artificial intelligence (AI) models using data from wearable biosensors to predict mental health conditions and symptoms. Following PRISMA guidelines, we identified 48 studies using a variety of wearable and smartphone biosensors including heart rate, heart rate variability (HRV), electrodermal activity/galvanic skin response (EDA/GSR), and digital proxies for biosignals such as accelerometry, location, audio, and usage metadata. We observed several technical and methodological challenges across studies in this field, including lack of ecological validity, data heterogeneity, small sample sizes, and battery drainage issues. We outline several corresponding opportunities for advancement in the field of AI-driven biosensing for mental health. Full article

Substance use disorders affect 17.3% of Americans. Digital health solutions that use machine learning to detect substance use from wearable biosignal data can eventually pave the way for real-time digital interventions. However, difficulties in addressing severe between-subject data heterogeneity have hampered the adaptation of machine learning approaches for substance use detection, necessitating more robust technological solutions. We tested the utility of personalized machine learning using participant-specific convolutional neural networks (CNNs) enhanced with self-supervised learning (SSL) to detect drug use. In a pilot feasibility study, we collected data from 9 participants using Fitbit Charge 5 devices, supplemented by ecological momentary assessments to collect real-time labels of substance use. We implemented a baseline 1D-CNN model with traditional supervised learning and an experimental SSL-enhanced model to improve individualized feature extraction under limited label conditions. Results: Among the 9 participants, we achieved an average area under the receiver operating characteristic curve score across participants of 0.695 for the supervised CNNs and 0.729 for the SSL models. Strategic selection of an optimal threshold enabled us to optimize either sensitivity or specificity while maintaining reasonable performance for the other metric. Conclusion: These findings suggest that Fitbit data have the potential to enhance substance use monitoring systems. However, the small sample size in this study limits its generalizability to diverse populations, so we call for future research that explores SSL-powered personalization at a larger scale. Full article

Hepatocellular carcinoma (HCC) has the highest mortality rate and is the most frequent of liver cancers. The heterogeneity of HCC in its etiology and molecular expression increases the difficulty in identifying possible treatments. To elucidate the molecular mechanisms of HCC across grades, data from The Cancer Genome Atlas (TCGA) were used for gene co-expression analysis, categorizing each sample into its pre-existing risk factors. The R library BioNERO was used for preprocessing and gene co-expression network construction. For those modules most correlated with a grade, functional enrichments from different databases were then tested, which appeared to have relatively consistent patterns when grouped by G1/G2 and G3/G4. G1/G2 exhibited the involvement of pathways related to metabolism and the PI3K/Akt pathway, which regulates cell proliferation and related pathways, whereas G3/G4 showed the activation of cell adhesion genes and the p53 signaling pathway, which regulates apoptosis, cell cycle arrest, and similar processes. Module preservation analysis was then used with the no history dataset as the reference network, which found cell adhesion molecules and cell cycle genes to be preserved across all risk factors, suggesting they are imperative in the development of HCC regardless of potential etiology. Through hierarchical clustering, modules related to the cell cycle, cell adhesion, the immune system, and the ribosome were found to be consistently present across all risk factors, with distinct clusters linked to oxidative phosphorylation in viral HCC and pentose and glucuronate interconversions in non-viral HCC, underscoring their potential roles in cancer progression. Full article

Diabetes mellitus is a heterogeneous metabolic disorder that poses significant health and economic challenges across the globe. Polysaccharides, found abundantly in edible plants, hold promise for managing diabetes by reducing blood glucose levels (BGL) and insulin resistance. However, most of these polysaccharides cannot be digested or absorbed directly by the human body. Here we report the production of antidiabetic oligosaccharides from cress seed mucilage polysaccharides using yeast fermentation. The water-soluble polysaccharides extracted from cress seed mucilage were precipitated using 75% ethanol and fermented with Pichia pastoris for different time intervals. The digested saccharides were fractionated through gel permeation chromatography using a Bio Gel P-10 column. Structural analysis of the oligosaccharide fractions revealed the presence of galacturonic acid, rhamnose, glucuronic acid, glucose and arabinose. Oligosaccharide fractions exhibited the potential to inhibit α-amylase and α-glucosidase enzymes in a dose-dependent manner in vitro. The fraction DF73 exhibited strong inhibitory activity against α-amylase with IC₅₀ values of 38.2 ± 1.12 µg/mL, compared to the positive control, acarbose, having an IC₅₀ value of 29.18 ± 1.76 µg/mL. Similarly, DF72 and DF73 showed the highest inhibition of α-glucosidase, with IC₅₀ values of 9.26 ± 2.68 and 50.47 ± 5.18 µg/mL, respectively. In in vivo assays in streptozotocin (STZ)-induced diabetic mice, these oligosaccharides significantly reduced BGL and improved lipid profiles compared to the reference drug metformin. Histopathological observations of mouse livers indicated the cytoprotective effects of these sugars. Taken together, our results suggest that oligosaccharides produced through microbial digestion of polysaccharides extracted from cress seed mucilage have the potential to reduce blood glucose levels, possibly through inhibition of carbohydrate-digesting enzymes and regulation of the various signaling pathways. Full article

We are living in an era of advanced nanoscience and nanotechnology. Numerous nanomaterials, culminating in nanorobots, have demonstrated ingenious applications in biomedicine, including breast cancer (BC) nano-theranostics. To solve the complicated problem of BC heterogeneity, non-targeted drug distribution, invasive diagnostics or surgery, resistance to classic onco-therapies and real-time monitoring of tumors, nanorobots are designed to perform multiple tasks at a small scale, even at the organelles or molecular level. Over the last few years, most nanorobots have been bioengineered as biomimetic and biocompatible nano(bio)structures, resembling different organisms and cells, such as urchin, spider, octopus, fish, spermatozoon, flagellar bacterium or helicoidal cyanobacterium. In this review, readers will be able to deepen their knowledge of the structure, behavior and role of several types of nanorobots, among other nanomaterials, in BC theranostics. We summarized here the characteristics of many functionalized nanodevices designed to counteract the main neoplastic hallmark features of BC, from sustaining proliferation and evading anti-growth signaling and resisting programmed cell death to inducing angiogenesis, activating invasion and metastasis, preventing genomic instability, avoiding immune destruction and deregulating autophagy. Most of these nanorobots function as targeted and self-propelled smart nano-carriers or nano-drug delivery systems (nano-DDSs), enhancing the efficiency and safety of chemo-, radio- or photodynamic therapy, or the current imagistic techniques used in BC diagnosis. Most of these nanorobots have been tested in vitro, using various BC cell lines, as well as in vivo, mainly based on mice models. We are still waiting for nanorobots that are low-cost, as well as for a wider transition of these favorable effects from laboratory to clinical practice. Full article

Liposarcomas (LPSs) are a heterogeneous group of malignancies that arise from adipose tissue. Although LPSs are among the most common soft-tissue sarcoma subtypes, precision medicine treatments are not currently available. To discover LPS-subtype-specific therapy targets, we investigated RNA sequenced transcriptomes of 131 clinical LPS tissue samples and compared the data with a transcriptome database that contained 20,218 samples from 95 healthy tissues and 106 cancerous tissue types. The identified genes were referred to the NCATS BioPlanet library with Enrichr to analyze upregulated signaling pathways. PDE3A protein expression was investigated with immunohistochemistry in 181 LPS samples, and PDE3A and SLFN12 mRNA expression with RT-qPCR were investigated in 63 LPS samples. Immunoblotting and cell viability assays were used to study LPS cell lines and their sensitivity to PDE3A modulators. We identified 97, 247, and 37 subtype-specific, highly expressed genes in dedifferentiated, myxoid, and pleomorphic LPS subtypes, respectively. Signaling pathway analysis revealed a highly activated hedgehog signaling pathway in dedifferentiated LPS, phospholipase c mediated cascade and insulin signaling in myxoid LPS, and pathways associated with cell proliferation in pleomorphic LPS. We discovered a strong association between high PDE3A expression and myxoid LPS, particularly in high-grade tumors. Moreover, myxoid LPS samples showed elevated expression levels of SLFN12 mRNA. In addition, PDE3A- and SLFN12-coexpressing LPS cell lines SA4 and GOT3 were sensitive to PDE3A modulators. Our results indicate that PDE3A modulators are promising drugs to treat myxoid LPS. Further studies are required to develop these drugs for clinical use. Full article

Stress is widely recognized as a major contributor to a variety of health issues. Stress prediction using biosignal data recorded by wearables is a key area of study in mobile sensing research because real-time stress prediction can enable digital interventions to immediately react at the onset of stress, helping to avoid many psychological and physiological symptoms such as heart rhythm irregularities. Electrodermal activity (EDA) is often used to measure stress. However, major challenges with the prediction of stress using machine learning include the subjectivity and sparseness of the labels, a large feature space, relatively few labels, and a complex nonlinear and subjective relationship between the features and outcomes. To tackle these issues, we examined the use of model personalization: training a separate stress prediction model for each user. To allow the neural network to learn the temporal dynamics of each individual’s baseline biosignal patterns, thus enabling personalization with very few labels, we pre-trained a one-dimensional convolutional neural network (1D CNN) using self-supervised learning (SSL). We evaluated our method using the Wearable Stress and Affect Detection(WESAD) dataset. We fine-tuned the pre-trained networks to the stress-prediction task and compared against equivalent models without any self-supervised pre-training. We discovered that embeddings learned using our pre-training method outperformed the supervised baselines with significantly fewer labeled data points: the models trained with SSL required less than 30% of the labels to reach equivalent performance without personalized SSL. This personalized learning method can enable precision health systems that are tailored to each subject and require few annotations by the end user, thus allowing for the mobile sensing of increasingly complex, heterogeneous, and subjective outcomes such as stress. Full article

Diffuse large B-cell lymphoma (DLBCL) is the most common form of non-Hodgkin lymphoma and frequently develops through the accumulation of several genetic variations. With the advancement in high-throughput techniques, in addition to mutations and copy number variations, structural variations have gained importance for their role in genome instability leading to tumorigenesis. In this study, in order to understand the genetics of DLBCL pathogenesis, we carried out a whole-genome mutation profile analysis of eleven human cell lines from germinal-center B-cell-like (GCB-7) and activated B-cell-like (ABC-4) subtypes of DLBCL. Analysis of genetic variations including small sequence variants and large structural variations across the cell lines revealed distinct variation profiles indicating the heterogeneous nature of DLBCL and the need for novel patient stratification methods to design potential intervention strategies. Validation and prognostic significance of the variants was assessed using annotations provided for DLBCL samples in cBioPortal for Cancer Genomics. Combining genetic variations revealed new subgroups between the subtypes and associated enriched pathways, viz., PI3K-AKT signaling, cell cycle, TGF-beta signaling, and WNT signaling. Mutation landscape analysis also revealed drug–variant associations and possible effectiveness of known and novel DLBCL treatments. From the whole-genome-based mutation analysis, our findings suggest putative molecular genetics of DLBCL lymphomagenesis and potential genomics-driven precision treatments. Full article

Humans are feeling emotions every day, but they can still encounter difficulties understanding them. To better understand emotions, we integrated interdisciplinary knowledge about emotions from various domains such as neurosciences (e.g., neurobiology), physiology, and psychology (affective sciences, positive psychology, cognitive psychology, psychophysiology, neuropsychology, etc.). To organize the knowledge, we employ technologies such as Artificial Intelligence with Knowledge Graphs and Semantic Reasoning. Furthermore, Internet of Things (IoT) technologies can help to acquire physiological data knowledge. The goal of this paper is to aggregate the interdisciplinary knowledge and implement it within the Emotional Knowledge Graph (EmoKG). The Emotional Knowledge Graph is used within our naturopathy recommender system that suggests food to boost emotion (e.g., chocolate contains magnesium that is recommended when we feel depressed). The recommender system also answers a set of competency questions to easily retrieve emotional related-knowledge from EmoKG, such as what are the basic emotions and the more sophisticated ones, what are the neurotransmitters and hormones related to emotions, etc. To follow FAIR principles, EmoKG is mapped to existing knowledge bases found on the BioPortal biomedical ontology catalog such as SNOMEDCT, FMA, RXNORM, MedDRA, and also from emotion ontologies (when available online). We design the LOV4IoT-Emotion ontology catalog that encourages researchers from heterogeneous communities to apply FAIR principles by releasing online their (emotion) ontologies, datasets, rules, etc. The set of ontology codes shared online can be semi-automatically processed; if not available, the scientific publications describing the emotion ontologies are semi-automatically processed with Natural Language Processing (NLP) technologies. This research is also relevant for other use cases such as European projects (ACCRA for emotional robots to reduce the social isolation of aging people, StandICT for standardization, and AI4EU for Artificial Intelligence) and alliances for IoT such as AIOTI. The recommender system can be extended to address other advice such as aromatherapy and take into consideration medical devices to monitor patients’ vital signals related to emotions and mental health. Full article

Surface-enhanced Raman spectroscopy (SERS) exploiting Raman reporter-labeled nanoparticles (RR@NPs) represents a powerful tool for the improvement of optical bio-assays due to RRs’ narrow peaks, SERS high sensitivity, and potential for multiplexing. In the present work, starting from low-cost and highly available raw materials such as cysteamine and substituted benzoic acids, novel bioorthogonal RRs, characterized by strong signal (10³ counts with FWHM < 15 cm⁻¹) in the biological Raman-silent region (>2000 cm⁻¹), RRs are synthesized by implementing a versatile, modular, and straightforward method with high yields and requiring three steps lasting 18 h, thus overcoming the limitations of current reported procedures. The resulting RRs’ chemical structure has SH-pendant groups exploited for covalent conjugation to high anisotropic gold-NPs. RR@NPs constructs work as SERS nanoprobes demonstrating high colloidal stability while retaining NPs’ physical and vibrational properties, with a limit of detection down to 60 pM. RR@NPs constructs expose carboxylic moieties for further self-assembling of biomolecules (such as antibodies), conferring tagging capabilities to the SERS nanoprobes even in heterogeneous samples, as demonstrated with in vitro experiments by transmembrane proteins tagging in cell cultures. Finally, thanks to their non-overlapping spectra, we envision and preliminary prove the possibility of exploiting RR@NPs constructs simultaneously, aiming at improving current SERS-based multiplexing bioassays. Full article

Focal onset epileptic seizures are highly heterogeneous in their clinical manifestations, and a robust seizure detection across patient cohorts has to date not been achieved. Here, we assess and discuss the potential of supervised machine learning models for the detection of focal onset motor seizures by means of a wrist-worn wearable device, both in a personalized context as well as across patients. Wearable data were recorded in-hospital from patients with epilepsy at two epilepsy centers. Accelerometry, electrodermal activity, and blood volume pulse data were processed and features for each of the biosignal modalities were calculated. Following a leave-one-out approach, a gradient tree boosting machine learning model was optimized and tested in an intra-subject and inter-subject evaluation. In total, 20 seizures from 9 patients were included and we report sensitivities of 67% to 100% and false alarm rates of down to 0.85 per 24 h in the individualized assessment. Conversely, for an inter-subject seizure detection methodology tested on an out-of-sample data set, an optimized model could only achieve a sensitivity of 75% at a false alarm rate of 13.4 per 24 h. We demonstrate that robustly detecting focal onset motor seizures with tonic or clonic movements from wearable data may be possible for individuals, depending on specific seizure manifestations. Full article

The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase whose proliferative effects can contribute to the development of many types of solid tumors when overexpressed. For this reason, EGFR inhibitors such as cetuximab can play an important role in treating cancers such as colorectal cancer and head and neck cancer. Cetuximab is a chimeric monoclonal antibody containing mouse variable regions that bind to EGFR and prevent it from signaling. Although cetuximab has been used clinically since 2004 to successfully control solid tumors, advances in protein engineering have created the opportunity to address some of its shortcomings. In particular, the presence of mouse sequences could contribute to immunogenicity in the form of anti-cetuximab antibodies, and an occupied glycosylation site in FR3 can contribute to hypersensitivity reactions and product heterogeneity. Using simple framework graft or sequence-/structure-guided approaches, cetuximab was humanized onto 11 new frameworks. In addition to increasing humanness and removing the VH glycosylation site, dynamic light scattering revealed increases in stability, and bio-layer interferometry confirmed minimal changes in binding affinity, with patterns emerging across the humanization method. This work demonstrates the potential to improve the biophysical and clinical properties of first-generation protein therapeutics and highlights the advantages of computationally guided engineering. Full article