dorsal/arxiv
View SchemaOligopeptides' frequencies in the classification of proteins' primary structures
| Authors | P. Sirabella, A. Giuliani, A. Colosimo |
|---|---|
| Categories | |
| ArXiv ID | physics/9807059 |
| URL | https://arxiv.org/abs/physics/9807059 |
Abstract
This paper reports about an approach to the classification of proteins' primary structures taking advantage of the Self Organizing Maps algorithm and of a numerical coding of the aminoacids based upon their physico-chemical properties. Hydrophobicity, volume, surface area, hydrophilicity, bulkiness, refractivity and polarity were subjected to a Principal Component Analysis and the first two principal components, explaining 84.8 % of the total observed variability, were used to cluster the aminoacids into 4 or 5 classes through a k-means algorithm. This leads to an economical representation of the primary structures which, in the construction of the input vectors for the Self Organizing Maps algorithm, allows the consideration of up to tri- and tetrapeptides' frequency matrices with minimal computational overload. In comparison with previously explored conditions, namely symbolic coding of aminoacids and dipeptides frequencies, no significant improvement was observed in the classification of 69 cytochromes of the c type, characterized by a high degree of structural and functional similarity, while a substantial improvement occurred in the case of a data set including quite heterogeneous primary structures.
{
"annotation_id": "bf606dd9-9b49-4d25-b2cc-ca550b68821e",
"date_created": "2026-03-02T18:01:21.524000Z",
"date_modified": "2026-03-02T18:01:21.524000Z",
"file_hash": "336ff17a227d63d94c4b329eaa4d672d12aab13c78bb156bfc4fa0dfc005a459",
"private": false,
"record": {
"abstract": "This paper reports about an approach to the classification of proteins\u0027\nprimary structures taking advantage of the Self Organizing Maps algorithm and\nof a numerical coding of the aminoacids based upon their physico-chemical\nproperties. Hydrophobicity, volume, surface area, hydrophilicity, bulkiness,\nrefractivity and polarity were subjected to a Principal Component Analysis and\nthe first two principal components, explaining 84.8 % of the total observed\nvariability, were used to cluster the aminoacids into 4 or 5 classes through a\nk-means algorithm. This leads to an economical representation of the primary\nstructures which, in the construction of the input vectors for the Self\nOrganizing Maps algorithm, allows the consideration of up to tri- and\ntetrapeptides\u0027 frequency matrices with minimal computational overload. In\ncomparison with previously explored conditions, namely symbolic coding of\naminoacids and dipeptides frequencies, no significant improvement was observed\nin the classification of 69 cytochromes of the c type, characterized by a high\ndegree of structural and functional similarity, while a substantial improvement\noccurred in the case of a data set including quite heterogeneous primary\nstructures.",
"arxiv_id": "physics/9807059",
"authors": [
"P. Sirabella",
"A. Giuliani",
"A. Colosimo"
],
"categories": [
"physics.bio-ph",
"physics.data-an",
"q-bio.QM"
],
"title": "Oligopeptides\u0027 frequencies in the classification of proteins\u0027 primary structures",
"url": "https://arxiv.org/abs/physics/9807059"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "30f41390-83a3-4400-8304-800221ba1d11",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}