Supplementary Materials: Pyrethroid Pesticide Metabolite in Urine and Microelements in Hair of Children Affected by Autism Spectrum Disorders: A Preliminary Investigation

“The Self-Organizing Map (SOM), commonly also known as Kohonen network [1,2] is a computational method for the visualization and analysis of high-dimensional data, especially experimentally acquired information.” Roughly speaking, SOMs projects high-dimensional data into a two-dimensional map. This procedure is called “mapping” and it preserves the topology of the data in a way that similar data items, or samples, will be mapped to nearby locations on the SOM [3]. The SOM is usually characterized by being a lattice, or grid, consisting of a grid of N2 nodes (also called neurons or cells) where N is the number of neurons for each side of the lattice. Nodes can be represented as rectangles or hexagons. Each grid node is connected to adjacent grid nodes according to a spatial neighborhood relation: a rectangular node has four neighbors while and hexagonal cell has six neighbors (Figure S1).