Biology subjects

Cardoso, A. S.

Publications and source records attributed to Cardoso, A. S..

2 recordsLinked to original sources

bioRxiv2021

Oral Lichen Planus and its relation with Oral Squamous Cell Carcinoma: new insights into the potential for malignant transformation

Oral Lichen Planus (OLP) is a chronic inflammatory disorder of unknown etiology. However, evidence suggests that it consists of an immunological process that leads to degeneration of the keratinocytes in the basal layer of the oral mucosa. Despite being recognized by WHO as a potentially malignant disorder with progression to oral squamous cell carcinoma (OSCC), the relationship between both pathologies is still controversial. Different studies have investigated factors associated with the potential for malignant transformation of OLP but it remains unclear. Through a bioinformatics approach, we investigated similarities in gene expression profiles of OLP and OSCC in early and advanced stages. Our results revealed gene expression patterns related to processes of keratinization, keratinocyte differentiation, cell proliferation and immune response in common between OLP and early and advanced OSCC, with the cornified envelope formation and antigen processing cross-presentation pathways in common between OLP and early OSCC. Together, these results reveal that key genes such as PI3, SPRR1B and KRT17, in addition to genes associated with different immune processes such as CXCL-13, HIF1A and IL1B may be involved in this oncogenic process. In addition, we performed an analysis of differentially and co-expressed genes and proposed putative therapeutic targets and associated drugs.

bioinformatics↗

bioRxiv2021

Deep learning assessment of cultural ecosystem services from social media images

Crowdsourced social media data has become popular in the assessment of cultural ecosystem services (CES). Advances in deep learning show great potential for the timely assessment of CES at large scales. Here, we describe a procedure for automating the assessment of image elements pertaining to CES from social media. We focus on a binary (natural, human) and a multiclass (posing, species, nature, landscape, human activities, human structures) classification of those elements using two Convolutional Neural Networks (CNNs; VGG16 and ResNet152) with the weights from two large datasets - Places365 and ImageNet -, and our own dataset. We train those CNNs over Flickr and Wikiloc images from the Peneda-Geres region (Portugal) and evaluate their transferability to wider areas, using Sierra Nevada (Spain) as test. CNNs trained for Peneda-Geres performed well, with results for the binary classification (F1-score > 80%) exceeding those for the multiclass classification (> 60%). CNNs pre-trained with Places365 and ImageNet data performed significantly better than with our data. Model performance decreased when transferred to Sierra Nevada, but their performances were satisfactory (> 60%). The combination of manual annotations, freely available CNNs and pre-trained local datasets thereby show great relevance to support automated CES assessments from social media.

ecology↗