PT  - JOURNAL ARTICLE
AU  - Paolo Inglese
AU  - Gonçalo Correia
AU  - Pamela Pruski
AU  - Robert C Glen
AU  - Zoltan Takats
TI  - Co-localization features for classification of tumors using mass spectrometry imaging
AID  - 10.1101/440057
DP  - 2018 Jan 01
TA  - bioRxiv
PG  - 440057
4099  - http://biorxiv.org/content/early/2018/10/11/440057.short
4100  - http://biorxiv.org/content/early/2018/10/11/440057.full
AB  - Statistical modeling of mass spectrometry imaging (MSI) data is a crucial component for the understanding of the molecular characteristics of cancerous tissues. Quantification of the abundances of metabolites or batch effect between multiple spectral acquisitions represents only a few of the challenges associated with this type of data analysis. Here we introduce a method based on ion co-localization features that allows the classification of whole tissue specimens using MSI data, which overcomes the possible batch effect issues and generates data-driven hypotheses on the underlying mechanisms associated with the different classes of analyzed samples.