Mutation-Attention (MuAt): deep representation learning of somatic mutations for tumour typing and subtyping.

Background: Cancer genome sequencing enables accurate classification of tumours and tumour subtypes. However, prediction performance is still limited using exome-only sequencing and for tumour types with low somatic mutation burden such as many paediatric tumours. Moreover, the ability to leverage deep representation learning in discovery of tumour entities remains unknown.

Cyclin B1 overexpression in conventional oral squamous cell carcinoma and verrucous carcinoma- A correlation with clinicopathological features.

Background: Nuclear localization of cyclin B1 is an indicator for cells undergoing mitotic division, and the overexpression has shown promising results as a good prognostic predictor for patients of squamous cell carcinoma (SCC). Cyclin B1 overexpression among histological grades of conventional oral squamous cell carcinoma (COSCC), as well as comparison with verrucous carcinoma (VC) has been less investigated.

Study Design: Immunohistochemical expression of cyclin B1 was compared with various clinicopathological features in 30 primary COSCC and 31 primary VC cases.

Candida in oral pre-cancer and oral cancer.

Oral pre-malignancies and carcinomas are common epithelial pathologies caused by a variety of etiological factors. In the oral cavity, candidiasis is the most frequent opportunistic fungal infection. Since the initial reports of an association between candidiasis with oral pre-cancer and cancer, various theories have been debated regarding the role of candida in development and transformation of oral pre-malignancies.