Machine Learning Techniques for the Prediction of phospholipidosis

Ph.D. student Mr. Robert Lowe.

Phospholipidosis was first believed to be observed by Nelson and Fitzhugh in 1948, when they reported the accumulation of foam macrophages in rats after long term treatment with chloroquine. It has been observed since then that numerous Cationic Amphiphilic Drugs can induce phospholipidosis in several cell types and that it can be characterized by the excess accumulation of phospholipids. Electron microscopy is the most reliable method of identifying whether a compound has induced phospholipidosis by the presence of lamellar inclusion bodies. It may also be identied by light microscopy in which cells appear vacuolated and contain foamy acrophages and vacuolation. The observation of compound-induced phospholipidosis in the drug design process is considered manageable as the effect often only occurs at very high doses, many times that of the intended therapeutic dose. There is also no strong evidence that that the condition is harmful to human health and it is reversible once treatment is terminated, the drug is expelled from the cell and phospholipid levels return to normal. This process can take weeks however and in some cases has been reported to last several months. The occurrence of phospholipidosis in the drug design process therefore can cause delays as more tests need to be carried out to satisfy regulatory bodies and perhaps sometimes stop the process altogether. The initial objectives are to create reliable models for phospholipidosis of xenobiotics using molecular properties and machine learning.

• ‹ previous
• 23 of 24
• next ›