Metabolomics analysis of pathways underlying radiation-induced salivary gland dysfunction stages.

Salivary gland hypofunction is an adverse side effect associated with radiotherapy for head and neck cancer patients. This study delineated metabolic changes at acute, intermediate, and chronic radiation damage response stages in mouse salivary glands following a single 5 Gy dose. Ultra-high performance liquid chromatography-mass spectrometry was performed on parotid salivary gland tissue collected at 3, 14, and 30 days following radiation (IR). Pathway enrichment analysis, network analysis based on metabolite structural similarity, and network analysis based on metabolite abundance correlations were used to incorporate both metabolite levels and structural annotation. The greatest number of enriched pathways are observed at 3 days and the lowest at 30 days following radiation. Amino acid metabolism pathways, glutathione metabolism, and central carbon metabolism in cancer are enriched at all radiation time points across different analytical methods. This study suggests that glutathione and central carbon metabolism in cancer may be important pathways in the unresolved effect of radiation treatment.

BRACETS: Bimodal repository of auscultation coupled with electrical impedance thoracic signals.

BACKGROUND AND OBJECTIVE: Respiratory diseases are among the most significant causes of morbidity and mortality worldwide, causing substantial strain on society and health systems. Over the last few decades, there has been increasing interest in the automatic analysis of respiratory sounds and electrical impedance tomography (EIT). Nevertheless, no publicly available databases with both respiratory sound and EIT data are available. METHODS: In this work, we have assembled the first open-access bimodal database focusing on the differential diagnosis of respiratory diseases (BRACETS: Bimodal Repository of Auscultation Coupled with Electrical Impedance Thoracic Signals). It includes simultaneous recordings of single and multi-channel respiratory sounds and EIT. Furthermore, we have proposed several machine learning-based baseline systems for automatically classifying respiratory diseases in six distinct evaluation tasks using respiratory sound and EIT (A1, A2, A3, B1, B2, B3). These tasks included classifying respiratory diseases at sample and subject levels. The performance of the classification models was evaluated using a 5-fold cross-validation scheme (with subject isolation between folds). RESULTS: The resulting database consists of 1097 respiratory sounds and 795 EIT recordings acquired from 78 adult subjects in two countries (Portugal and Greece). In the task of automatically classifying respiratory diseases, the baseline classification models have achieved the following average balanced accuracy: Task A1 - 77.9±13.1%; Task A2 - 51.6±9.7%; Task A3 - 38.6±13.1%; Task B1 - 90.0±22.4%; Task B2 - 61.4±11.8%; Task B3 - 50.8±10.6%. CONCLUSION: The creation of this database and its public release will aid the research community in developing automated methodologies to assess and monitor respiratory function, and it might serve as a benchmark in the field of digital medicine for managing respiratory diseases. Moreover, it could pave the way for creating multi-modal robust approaches for that same purpose.

Automatic wheeze segmentation using harmonic-percussive source separation and empirical mode decomposition.

Wheezes are adventitious respiratory sounds commonly present in patients with respiratory conditions. The presence of wheezes and their time location are relevant for clinical reasons, such as understanding the degree of bronchial obstruction. Conventional auscultation is usually employed to analyze wheezes, but remote monitoring has become a pressing need during recent years. Automatic respiratory sound analysis is required to reliably perform remote auscultation. In this work we propose a method for wheeze segmentation. Our method starts by decomposing a given audio excerpt into intrinsic mode frequencies using empirical mode decomposition. Then, we apply harmonic-percussive source separation to the resulting audio tracks and get harmonic-enhanced spectrograms, which are processed to obtain harmonic masks. Subsequently, a series of empirically derived rules are applied to find wheeze candidates. Finally, the candidates stemming from the different audio tracks are merged and median filtered. In the evaluation stage, we compare our method to three baselines on the ICBHI 2017 Respiratory Sound Database, a challenging dataset containing various noise sources and background sounds. Using the full dataset, our method outperforms the baselines, achieving an F1 of 41.9%. Our method's performance is also better than the baselines across several stratified results focusing on five variables: recording equipment, age, sex, body-mass index, and diagnosis. We conclude that, contrary to what has been reported in the literature, wheeze segmentation has not been solved for real life scenario applications. Adaptation of existing systems to demographic characteristics might be a promising step in the direction of algorithm personalization, which would make automatic wheeze segmentation methods clinically viable.

Automatic segmentation and classification of mice ultrasonic vocalizations.

This paper addresses the development of a system for classifying mouse ultrasonic vocalizations (USVs) present in audio recordings. The automatic labeling process for USVs is usually divided into two main steps: USV segmentation followed by the matching classification. Three main contributions can be highlighted: (i) a new segmentation algorithm, (ii) a new set of features, and (iii) the discrimination of a higher number of classes when compared to similar studies. The developed segmentation algorithm is based on spectral entropy analysis. This novel segmentation approach can detect USVs with 94% and 74% recall and precision, respectively. When compared to other methods/software, our segmentation algorithm achieves a higher recall. Regarding the classification phase, besides the traditional features from time, frequency, and time-frequency domains, a new set of contour-based features were extracted and used as inputs of shallow machine learning classification models. The contour-based features were obtained from the time-frequency ridge representation of USVs. The classification methods can differentiate among ten different syllable types with 81.1% accuracy and 80.5% weighted F1-score. The algorithms were developed and evaluated based on a large dataset, acquired on diverse social interaction conditions between the animals, to stimulate a varied vocal repertoire.

Annotation and functional characterization of long noncoding RNAs deregulated in pancreatic adenocarcinoma.

PURPOSE: Transcriptome analysis of pancreatic ductal adenocarcinoma (PDAC) has been useful to identify gene expression changes that sustain malignant phenotypes. Yet, most studies examined only tumor tissues and focused on protein-coding genes, leaving long non-coding RNAs (lncRNAs) largely underexplored. METHODS: We generated total RNA-Seq data from patient-matched tumor and nonmalignant pancreatic tissues and implemented a computational pipeline to survey known and novel lncRNAs. siRNA-mediated knockdown in tumor cell lines was performed to assess the contribution of PDAC-associated lncRNAs to malignant phenotypes. Gene co-expression network and functional enrichment analyses were used to assign deregulated lncRNAs to biological processes and molecular pathways. RESULTS: We detected 9,032 GENCODE lncRNAs as well as 523 unannotated lncRNAs, including transcripts significantly associated with patient outcome. Aberrant expression of a subset of novel and known lncRNAs was confirmed in patient samples and cell lines. siRNA-mediated knockdown of a subset of these lncRNAs (LINC01559, LINC01133, CCAT1, LINC00920 and UCA1) reduced cell proliferation, migration and invasion. Gene co-expression network analysis associated PDAC-deregulated lncRNAs with diverse biological processes, such as cell adhesion, protein glycosylation and DNA repair. Furthermore, UCA1 knockdown was shown to specifically deregulate co-expressed genes involved in DNA repair and to negatively impact DNA repair following damage induced by ionizing radiation. CONCLUSIONS: Our study expands the repertoire of lncRNAs deregulated in PDAC, thereby revealing novel candidate biomarkers for patient risk stratification. It also provides a roadmap for functional assays aimed to characterize novel mechanisms of action of lncRNAs in pancreatic cancer, which could be explored for therapeutic development.

Genome-wide promoter methylation profiling in a cellular model of melanoma progression reveals markers of malignancy and metastasis that predict melanoma survival.

The epigenetic changes associated with melanoma progression to advanced and metastatic stages are still poorly understood. To shed light on the CpG methylation dynamics during melanoma development, we analyzed the methylome profiles of a four-stage cell line model of melanoma progression: non-tumorigenic melanocytes (melan-a), premalignant melanocytes (4C), non-metastatic melanoma cells (4C11-), and metastatic melanoma cells (4C11+). We identified 540 hypo- and 37 hypermethylated gene promoters that together characterized a malignancy signature, and 646 hypo- and 520 hypermethylated promoters that distinguished a metastasis signature. Differentially methylated genes from these signatures were correlated with overall survival using TCGA-SKCM methylation data. Moreover, multivariate Cox analyses with LASSO regularization identified panels of 33 and 31 CpGs, respectively, from the malignancy and metastasis signatures that predicted poor survival. We found a concordant relationship between DNA methylation and transcriptional levels for genes from the malignancy (Pyroxd2 and Ptgfrn) and metastasis (Arnt2, Igfbp4 and Ptprf) signatures, which were both also correlated with melanoma prognosis. Altogether, this study reveals novel CpGs methylation markers associated with malignancy and metastasis that collectively could improve the survival prediction of melanoma patients.

GLUT3/SLC2A3 Is an Endogenous Marker of Hypoxia in Prostate Cancer Cell Lines and Patient-Derived Xenograft Tumors.

The microenvironment of solid tumors is dynamic and frequently contains pockets of low oxygen levels (hypoxia) surrounded by oxygenated tissue. Indeed, a compromised vasculature is a hallmark of the tumor microenvironment, creating both spatial gradients and temporal variability in oxygen availability. Notably, hypoxia associates with increased metastasis and poor survival in patients. Therefore, to aid therapeutic decisions and better understand hypoxia's role in cancer progression, it is critical to identify endogenous biomarkers of hypoxia to spatially phenotype oncogenic lesions in human tissue, whether precancerous, benign, or malignant. Here, we characterize the glucose transporter GLUT3/SLC2A3 as a biomarker of hypoxic prostate epithelial cells and prostate tumors. Transcriptomic analyses of non-tumorigenic, immortalized prostate epithelial cells revealed a highly significant increase in GLUT3 expression under hypoxia. Additionally, GLUT3 protein increased 2.4-fold in cultured hypoxic prostate cell lines and was upregulated within hypoxic regions of xenograft tumors, including two patient-derived xenografts (PDX). Finally, GLUT3 out-performs other established hypoxia markers; GLUT3 staining in PDX specimens detects 2.6-8.3 times more tumor area compared to a mixture of GLUT1 and CA9 antibodies. Therefore, given the heterogeneous nature of tumors, we propose adding GLUT3 to immunostaining panels when trying to detect hypoxic regions in prostate samples.

Genes regulated by DNA methylation are involved in distinct phenotypes during melanoma progression and are prognostic factors for patients.

In addition to mutations, epigenetic alterations are important contributors to malignant transformation and tumor progression. The aim of this work was to identify epigenetic events in which promoter or gene body DNA methylation induces gene expression changes that drive melanocyte malignant transformation and metastasis. We previously developed a linear mouse model of melanoma progression consisting of spontaneously immortalized melanocytes, premalignant melanocytes, a nonmetastatic tumorigenic, and a metastatic cell line. Here, through the integrative analysis of methylome and transcriptome data, we identified the relationship between promoter and/or gene body DNA methylation alterations and gene expression in early, intermediate, and late stages of melanoma progression. We identified adenylate cyclase type 3 (Adcy3) and inositol polyphosphate 4-phosphatase type II (Inpp4b), which affect tumor growth and metastatic potential, respectively. Importantly, the gene expression and DNA methylation profiles found in this murine model of melanoma progression were correlated with available clinical data from large population-based primary melanoma cohorts, revealing potential prognostic markers.

Classification of Electrical Impedance Tomography Data Using Machine Learning.

Patients suffering from pulmonary diseases typically exhibit pathological lung ventilation in terms of homogeneity. Electrical Impedance Tomography (EIT) is a non- invasive imaging method that allows to analyze and quantify the distribution of ventilation in the lungs. In this article, we present a new approach to promote the use of EIT data and the implementation of new clinical applications for differential diagnosis, with the development of several machine learning models to discriminate between EIT data from healthy and nonhealthy subjects. EIT data from 16 subjects were acquired: 5 healthy and 11 non-healthy subjects (with multiple pulmonary conditions). Preliminary results have shown accuracy percentages of 66% in challenging evaluation scenarios. The results suggest that the pairing of EIT feature engineering methods with machine learning methods could be further explored and applied in the diagnostic and monitoring of patients suffering from lung diseases. Also, we introduce the use of a new feature in the context of EIT data analysis (Impedance Curve Correlation).

Detection of squawks in respiratory sounds of mechanically ventilated COVID-19 patients.

Mechanically ventilated patients typically exhibit abnormal respiratory sounds. Squawks are short inspiratory adventitious sounds that may occur in patients with pneumonia, such as COVID-19 patients. In this work we devised a method for squawk detection in mechanically ventilated patients by developing algorithms for respiratory cycle estimation, squawk candidate identification, feature extraction, and clustering. The best classifier reached an F1 of 0.48 at the sound file level and an F1 of 0.66 at the recording session level. These preliminary results are promising, as they were obtained in noisy environments. This method will give health professionals a new feature to assess the potential deterioration of critically ill patients.

Transcriptional signatures underlying dynamic phenotypic switching and novel disease biomarkers in a linear cellular model of melanoma progression.

Despite advances in therapeutics, the progression of melanoma to metastasis still confers a poor outcome to patients. Nevertheless, there is a scarcity of biological models to understand cellular and molecular changes taking place along disease progression. Here, we characterized the transcriptome profiles of a multi-stage murine model of melanoma progression comprising a nontumorigenic melanocyte lineage (melan-a), premalignant melanocytes (4C), nonmetastatic (4C11-) and metastasis-prone (4C11+) melanoma cells. Clustering analyses have grouped the 4 cell lines according to their differentiated (melan-a and 4C11+) or undifferentiated/"mesenchymal-like" (4C and 4C11-) morphologies, suggesting dynamic gene expression patterns associated with the transition between these phenotypes. The cell plasticity observed in the murine melanoma progression model was corroborated by molecular markers described during stepwise human melanoma differentiation, as the differentiated cell lines in our model exhibit upregulation of transitory and melanocytic markers, whereas "mesenchymal-like" cells show increased expression of undifferentiated and neural crest-like markers. Sets of differentially expressed genes (DEGs) were detected at each transition step of tumor progression, and transcriptional signatures related to malignancy, metastasis and epithelial-to-mesenchymal transition were identified. Finally, DEGs were mapped to their human orthologs and evaluated in uni- and multivariate survival analyses using gene expression and clinical data of 703 drug-naïve primary melanoma patients, revealing several independent candidate prognostic markers. Altogether, these results provide novel insights into the molecular mechanisms underlying the phenotypic switch taking place during melanoma progression, reveal potential drug targets and prognostic biomarkers, and corroborate the translational relevance of this unique sequential model of melanoma progression.

Integration of metabolomics and transcriptomics reveals convergent pathways driving radiation-induced salivary gland dysfunction.

Radiation therapy for head and neck cancer causes damage to the surrounding salivary glands, resulting in salivary gland hypofunction and xerostomia. Current treatments do not provide lasting restoration of salivary gland function following radiation; therefore, a new mechanistic understanding of the radiation-induced damage response is necessary for identifying therapeutic targets. The purpose of the present study was to investigate the metabolic phenotype of radiation-induced damage in parotid salivary glands by integrating transcriptomic and metabolomic data. Integrated data were then analyzed to identify significant gene-metabolite interactions. Mice received a single 5 Gy dose of targeted head and neck radiation. Parotid tissue samples were collected 5 days following treatment for RNA sequencing and metabolomics analysis. Altered metabolites and transcripts significantly converged on a specific region in the metabolic reaction network. Both integrative pathway enrichment using rank-based statistics and network analysis highlighted significantly coordinated changes in glutathione metabolism, energy metabolism (TCA cycle and thermogenesis), peroxisomal lipid metabolism, and bile acid production with radiation. Integrated changes observed in energy metabolism suggest that radiation induces a mitochondrial dysfunction phenotype. These findings validated previous pathways involved in the radiation-damage response, such as altered energy metabolism, and identified robust signatures in salivary glands, such as reduced glutathione metabolism, that may be driving salivary gland dysfunction.

Automatic Classification of Adventitious Respiratory Sounds: A (Un)Solved Problem?

(1) Background: Patients with respiratory conditions typically exhibit adventitious respiratory sounds (ARS), such as wheezes and crackles. ARS events have variable duration. In this work we studied the influence of event duration on automatic ARS classification, namely, how the creation of the Other class (negative class) affected the classifiers' performance. (2) Methods: We conducted a set of experiments where we varied the durations of the other events on three tasks: crackle vs. wheeze vs. other (3 Class); crackle vs. other (2 Class Crackles); and wheeze vs. other (2 Class Wheezes). Four classifiers (linear discriminant analysis, support vector machines, boosted trees, and convolutional neural networks) were evaluated on those tasks using an open access respiratory sound database. (3) Results: While on the 3 Class task with fixed durations, the best classifier achieved an accuracy of 96.9%, the same classifier reached an accuracy of 81.8% on the more realistic 3 Class task with variable durations. (4) Conclusion: These results demonstrate the importance of experimental design on the assessment of the performance of automatic ARS classification algorithms. Furthermore, they also indicate, unlike what is stated in the literature, that the automatic classification of ARS is not a solved problem, as the algorithms' performance decreases substantially under complex evaluation scenarios.

A Wearable Stethoscope for Long-Term Ambulatory Respiratory Health Monitoring.

Lung sounds acquired by stethoscopes are extensively used in diagnosing and differentiating respiratory diseases. Although an extensive know-how has been built to interpret these sounds and identify diseases associated with certain patterns, its effective use is limited to individual experience of practitioners. This user-dependency manifests itself as a factor impeding the digital transformation of this valuable diagnostic tool, which can improve patient outcomes by continuous long-term respiratory monitoring under real-life conditions. Particularly patients suffering from respiratory diseases with progressive nature, such as chronic obstructive pulmonary diseases, are expected to benefit from long-term monitoring. Recently, the COVID-19 pandemic has also shown the lack of respiratory monitoring systems which are ready to deploy in operational conditions while requiring minimal patient education. To address particularly the latter subject, in this article, we present a sound acquisition module which can be integrated into a dedicated garment; thus, minimizing the role of the patient for positioning the stethoscope and applying the appropriate pressure. We have implemented a diaphragm-less acousto-electric transducer by stacking a silicone rubber and a piezoelectric film to capture thoracic sounds with minimum attenuation. Furthermore, we benchmarked our device with an electronic stethoscope widely used in clinical practice to quantify its performance.

Oxidative instability of boronic acid-installed polycarbonate nanoparticles.

Oxidative stress, caused by the overproduction of reactive oxygen species (ROS), is often observed in degenerative and/or metabolic diseases, tumors, and inflamed tissues. Boronic acids are emerging as a unique class of responsive biomaterials targeting ROS because of their reactivity toward H2O2. Herein, we examine the oxidative reactivity of nanoparticles from a boronic acid-installed polycarbonate. The extent of oxidation under different concentrations of H2O2 was tracked by the change in fluorescence intensity of an encapsulated solvatochromic reporter dye, demonstrating their sensitivity to biologically-relevant concentrations of hydrogen peroxide. Oxidation-triggered particle destabilization, however, was shown to be highly dependent on the concentration of the final oxidized polymer product, and was only achieved if it fell below polymer critical micelle concentration. Our results indicate that these nanocarriers serve as an excellent dual pH/H2O2 responsive vehicle for drug delivery.

Neutrophil extracellular traps possess anti-human respiratory syncytial virus activity: Possible interaction with the viral F protein.

Human respiratory syncytial virus (hRSV) is one of the main etiological agents of diseases of the lower respiratory tract, and is often responsible for the hospitalization of children and the elderly. To date, treatments are only palliative and there is no vaccine available. The airways of patients infected with hRSV exhibit intense neutrophil infiltration, which is responsible for the release of neutrophil extracellular traps (NETs). These are extracellular structures consisting of DNA associated with intracellular proteins, and are efficient in capturing and eliminating various microorganisms, including some viruses. hRSV induces the release of NETs into the lung tissue of infected individuals; however, the pathophysiological consequences of this event have not been elucidated. The objective of this study was to utilize in vitro and in silico assays to investigate the impact of NETs on hRSV infection. NETs, generated by neutrophils stimulated with phorbol myristate acetate (PMA), displayed long fragments of DNA and an electrophoretic profile suggestive of the presence of proteins that are classically associated with these structures (elastase, cathepsin G, myeloperoxidase, and histones). The presence of NETs (>2 µg/ml) in HEp-2 cell culture medium resulted in cellular cytotoxicity of less than 50%. Pre-incubation (1 h) of viral particles (multiplicity of infection (MOI) values of 0.1, 0.5, and 1.0) with NETs (2-32 µg/ml) resulted in cellular protection from virus-induced death of HEp-2 cells. Concurrently, there was a reduction in the formation of syncytia, which is related to decreased viral spread in infected tissue. Results from western blotting and molecular docking, suggest interactions between F protein of the hRSV viral envelope and BPI (bactericidal permeability-increasing protein), a microbicidal member of NETs. Interactions occurred at sites important for the neutralization and coordination of the hRSV infection/replication process. Our results showed that the presence of NETs decreases hRSV-induced cellular damage, possibly by directly affecting viral particle capture and/or interfering with the fusion activity of the F protein. These findings broaden the understanding of the role of NETs during hRSV infection.

Identificação e anotação funcional de RNAs longos não codificadores associados à subtipos moleculares em tumores pancreáticos / Identification of long non coding RNAs associated to molecular subtypes in pancreatic tumors

O Adenocarcinoma Pancreático Ductal (Pancreatic Ductal Adenocarcinoma - PDAC) é a sé- tima causa de mortes por câncer no mundo, com uma taxa de sobrevida de apenas 6%. Embora alguns genes estejam recorrentemente mutados em grande parte dos tumores e sejam críticos para a oncogênese, a heterogeneidade das alterações moleculares tanto no tumor quanto em componentes do microambiente tumoral se reflete em diferentes características fenotípicas com comportamentos clínicos distintos e que têm sido associados a diferentes subtipos moleculares através da análise computacional de dados de alterações somáticas e transcricionais no PDAC. RNAs não codificadores longos (lncRNAs) têm sido reconhecidos como importantes reguladores da expressão gênica em doenças proliferativas mas sua associação com subtipos em PDAC e sua contribuição para o estabelecimento de diferentes fenótipos moleculares e clínicos da doença não foi explorada até o momento. Neste trabalho, foi implementada uma abordagem computacional com o objetivo de identificar e anotar funcionalmente lncRNAs associados a subtipos moleculares de PDAC. Inicialmente, a classificação não supervisionada por Fatoração Matricial Não Negativa (Non-Negative Matrix Factorization - NMF) de dados de expressão gê- nica global de amostras clínicas disponíveis publicamente (The Cancer Genome Atlas - TCGA) resultou na identificação de quatro subgrupos distintos de PDAC, que recapitulam os fenótipos Exócrino/Endócrino, Imunogênico, Escamoso e Progenitor descritos na literatura. Uma análise de expressão diferencial permitiu a identificação de assinaturas de expressão gênica características que incluem lncRNAs associados a cada subgrupo. Através da construção de redes de coexpressão de mRNAs e lncRNAs e a identificação de módulos da rede significativamente enriquecidos em genes que participam em vias moleculares conhecidas foi possível inferir possíveis funções biológicas à lncRNAs associados aos diferentes subtipos moleculares, tais como funções exócrinas/neuroendócrinas, imunogênicas, reparo de DNA/progressão do ciclo celular e progenitoras/morfogênicas. Entre ele, o subgrupo 3, enriquecido para fenótipo Escamoso e associado a hiper-expressão do supressor tumoral TP63, possui dois lncRNAS hiper-expressos neste subgrupo em relação aos outros subgrupos, sendo que o lncRNA antissenso FAM83A-AS1 tem a predição de interagir com as proteínas FGFR2, AXIN1, PTEN, BRAF, SMAD4, TGFBR2, TP53 e CDKN2A, que exercem funções importantes na transdução de sinal e supressão tumoral no câncer incluindo o de pâncreas. Entre os lncRNAs hipo-regulados no subgrupo 3 em relação ao outros subgrupos, alguns, como FLJ42875, LOC338651, C20orf56 e LOC38838 tem predição de interação com alta afinidade à proteína BRCA2, que está envolvida no reparo de DNA e participa de processos de resistência à quimioterápicos. As informações trazidas por este estudo permitem gerar hipóteses sobre a contribuição de lncRNAs para a definição de subtipos moleculares de PDAC e priorizar candidatos e experimentos para estudos funcionais de modo a contribuir para um melhor entendimento sobre os mecanismos de ação de lncRNAs na tumorigênese e agressividade do câncer de pâncreas

Pancreatic Ductal Adenocarcinoma (PDAC) is the seventh cause of worldwide cancer related deaths, with an overall survival rate of only 6%. Some genes might be recurrently mutated in a large number of tumors, and be critical for oncogenesis, molecular alteration heterogeneity both in the tumor as all as in the tumor microenvironment is reflected in diverse phenotypic features with distinct clinical outcomes, and this distinction in multiple molecular subtypes has been drawn through transcriptional and somatic alteration computational analysis within PDAC. Long Non Coding RNAs (lncRNAs) have been recognized as important gene expression regulators in proliferative diseases, but its association to molecular subtypes in PDAC and its contribution in the establishment of diverse molecular and clinical phenotypes hasnt been explored at length until the present. This work focused on the implementation of a computational approach with the objective of lncRNA identification and functional annotation associated to distinct molecular subtypes in PDAC. Initially, Non-negative Matrix Factorization (NMF), an unsupervised classification method, applied to global gene expression data from publicly available clinical samples (The Cancer Genome Atlas - TCGA) resulted in the identification of four distinct PDAC molecular subgroups reminiscent of Exocrine/Endocrine, Immunogenic, Squamous and Progenitor phenotypes. Differential expression analysis allowed a characteristic gene expression signature identification, including distinct molecular subtype associated lncRNAs. mRNA and lncRNA containing gene co-expression modules significantly enriched annotated pathways containing the molecular subtype associated lncRNAs allowed to designate possible molecular functions of the distinct molecular subtype associated lncRNAs, such as exocrine/neuroendocrine, immunogenic, DNA repair/cell cycle progression and progenitor/morphogenic functions. Subgroup 3, enriched with a Squamous phenotype and associated to TP63 over-expression contains two lncRNAs over-expressed compared to other subgroups; furthermore, the antisense lncRNA FAM83A-AS1 yielded a predicted lncRNA-protein interaction to FGFR2, AXIN1, PTEN, BRAF, SMAD4, TGFBR2, TP53 and CDKN2A, proteins that play important signal transduction and tumor suppressor roles in several cancer types, including pancreas. Among under-expressed lncRNAs in subgroup 3 compared to the other subgroups, some, such as FLJ42875, LOC338651, C20orf56 and LOC38838 yilded a high protein interaction prediction score with BRCA2, a protein involved in DNA repair and processes resuling in chemotherapy resistance. The information brought by this study allowed to generate hypothesis on lncRNA contribution to define PDAC molecular subtypes, helping prioritize candidates and experiments for functional studies, thus contributing to a better understanding on lncRNA mechanisms related to tumor progression and aggressiveness in pancreatic cancer