Skip to main content
  • Oral presentation
  • Open access
  • Published:

The use of self-organizing maps to study fatty acids in neuropsychiatric disorders

In our study we have evaluated the theme of the platelet fatty acid composition in subjects with a clinical diagnosis of Major Depression (MD). We have investigated 84 subjects (51 females and 33 males, mean age: 60.21, SD: ± 12), compared with 60 control, apparently healthy subjects (38 Males and 22 Females, mean age: 33.97, SD: ± 12.40). The tools used for the diagnosis of Major Depression were: Clinical Global Impression (CGI), Symptoms Check List-90 (SCL-90), Medical and Pharmacological history, BMI, Structured Clinical Interview DSM-IV-SCID-IV (American Psychiatric Association 2000), Hamilton Rating Scale of Depression (HRSD).

We have analysed the groups without taking in account therapies, gender and age. The results obtained show the evidence of three fatty acids, Arachidonic Acid (AA), Linoleic Acid (LA), and Palmitic Acid (PA) in a peculiar position with respect to the biochemical characterization of MD. The ratio among the three fatty acids, in the different conditions studied, allows us to do the hypothesis that the MD is linked to a possible un-balance of the membrane function. The depressive condition, according to the fatty acid composition, is characterized by a progressive increase of the degree of un-saturation of the platelet fatty acids in most of the patients, while a consistent group is characterized by an increase of the degree of saturation. These findings in platelets seem to be more specific indicators with respect to the red blood cell fatty acids and plasma phospholipid fatty acids.

In conclusion, from the study, we can assume:

1) AA, LA and PA are the three main lipid biochemical markers in platelets that characterize the MD.

2) The constant sum of the three fatty acids (AA+LA+PA = 53.33 ± 3.43) in all the conditions studied other than the depressive and normal people, gives reason of the role that they play in the membrane balance. In particular the coefficient of correlation between the sum of AA and LA versus PA was highly significant: r= −.66 (p=0.00)

3) The differences of the degrees of un-saturation and of saturation in the platelets of the subjects could be a signal of a different condition within the same pathology. The use of the Artificial Neural Network (ANN) to analyse the modifications of the fatty acids allows us to confirm, from a biochemical point of view, the clinical diagnosis of MD.

4) It is possible that the blood platelets fatty acids (AA, LA, PA) are strong specific markers also in other psychiatric conditions and that they are an easy way to forecast, at least the MD, possibly, other pathologies.

Author information

Authors and Affiliations


Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution 2.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Cocchi, M., Tonello, L. & Cappello, G. The use of self-organizing maps to study fatty acids in neuropsychiatric disorders. Ann Gen Psychiatry 7 (Suppl 1), S84 (2008).

Download citation

  • Published:

  • DOI: