[Nov 28th 2018] LabMeeting: Novel Neural Techniques for Gene Expression Analysis in Cancer Prognosis

Gabriele Ciravegna (DIISM, University of Siena)

Nov 28, 2018 – 11:00 AM
DIISM, Artificial Intelligence laboratory (room 201), Siena SI
Description

Cancer is a large family of genetic diseases that involve abnormal cell growth. Genetic mutations can vary from one patient to another. Therefore, personalized precision is required to increase the reliability of prognostic predictions and the benefit of therapeutic decisions. The most important issues concerning gene analysis are strong noise and high dimensionality of data. In order to address these problems, a novel neural network for hierarchical clustering has been devised (the GH-EXIN). A benchmark comparison stated how the GH-EXIN neural network produces better data quantization with respect to SoA, while providing innovative capabilities as the outlier detection. The GH-EXIN is inserted within a biclustering framework, which allows to overcome the high dimensionality problem. Furthermore, biclustering allows to find those activation patterns which are common to a group of genes only under specific experimental conditions. Indeed, subsets of genes are coregulated and coexpressed only under certain experimental conditions, but behave almost independently under other conditions. Results of these analysis were presented in [1][2][3].
This biclustering framework was successfully tested also on an external dataset for face detection problem and the results were presented in [4].

[1] P. Barbiero, A. Bertotti, G. Ciravegna, G. Cirrincione, and E. Piccolo, “Neural biclustering in gene expression analysis”. In 2017 International Conference on Computational Science and Computational Intelligence (CSCI), 11 2017.
[2] P. Barbiero, A. Bertotti, G. Ciravegna, G. Cirrincione , E. Pasero and E. Piccolo, “Unsupervised Gene Identification in Colorectal Cancer”, Italian Workshop on Neural Networks (WIRN 2017), June 14- 16, 2017, Vietri sul Mare (SA), Italy.
[3] P. Barbiero, A. Bertotti, G. Ciravegna, G. Cirrincione , E. Pasero and E. Piccolo, “Supervised Gene Identification in Colorectal Cancer”, Italian Workshop on Neural Networks (WIRN 2017), June 14- 16, 2017, Vietri sul Mare (SA), Italy
[4] G. Ciravegna*, G. Cirrincione, F. Marcolin, P. Barbiero, N. Dagnes, and E. Piccolo, “Assessing discriminating capability of geometrical descriptors for 3d face recognition by using the GH-EXIN neural network.”, Italian Workshop on Neural Networks (WIRN 2018), June 14- 16, 2018, Vietri sul Mare (SA), Italy

 |  Category: Seminars