Online Inquiry

Please submit a detailed description of your project. We will provide you with a customized project plan to meet your research requests. You can also send emails directly to for inquiries.
Please input "proteomics" as verification code.
Related Services

SOM Service

A self-organizing map (SOM), also called self-organizing feature map (SOFM),is a type of artificial neural network(ANN) that is trained using unsupervised learning method to generate a low-dimensional, discretized representation (a map) of the input space of the training samples. SOMs differ from other ANN methods as they make use of competitive learning as opposed to error-correction learning, and in the sense that they make use of a neighborhood function to preserve the topological properties of the input space. This makes SOMs useful for visualizing low-dimensional views of high-dimensional data, similar to multidimensional scaling. Now, bioinformaticians at Creative Proteomics are proud to tell you we are open to help you with SOM Service!

SOM Service

Like most ANNs, SOMs function in two modes: training and mapping. "Training" constructs the map using input examples (a competitive process, also called vector quantization), while "mapping" automatically classifies a new input vector.A self-organizing map is composed of components called nodes or neurons. Associated with each node is a weight vector of the same dimension as the input data vectors, and a position in the map space. The typical arrangement of nodes is a two-dimensional ordered spacing in a hexagonal or rectangular grid. The SOM describes a mapping from a higher-dimensional input space to a lower-dimensional map space. The purpose for placing a vector from data space onto the map is to seek the node with the closest (smallest distance metric) weight vector to the data space vector.

The steps of the SOM algorithm can be summarized as follows:

  • Initialization: choose random values for the initial weight vectors wj.
  • Sampling: draw a sample training input vector x from the input data.
  • Matching: find the winning neuron I(x) with weight vector closest to input vector.
  • Updating: adopt the weight update equation ∆wji= η(t)Tj,I(x)(t) (xi − wji).
  • Continuation: keep returning to step 2 until the feature map stops changing.

How to place an order:
How To Place your Order

*If your organization requires signing of a confidentiality agreement, please contact us by email

Our customer service representatives are available 24 hours a day, from Monday to Sunday. Contact Us INQUIRY