<b>Human Colon Cancer Drug Release Nanoparticle Systems Design: </b><b>Dataset Compilation and Machine Learning Modeling</b>

  1. Munteanu, Cristian R.
  2. He, Shan
  3. de Bilbao, Begoña
  4. Bediaga, Harbil
  5. Casanola-Martin, Gerardo M.
  6. Ascencio, Estefania
  7. Chelu, Mariana
  8. Magdalena Musuc, Adina
  9. Arrasate, Sonia
  10. Pazos Sierra, Alejandro
  11. Rasulev, Bakhtiyor
  12. González-Díaz, Humberto

Argitaratzaile: figshare

Argitalpen urtea: 2024

Mota: Dataset

CC BY 4.0

DOI: 10.6084/M9.FIGSHARE.25983757 lock_openSarbide irekia editor

Laburpena

Nanoparticles (NPs) are interesting for Human Colon Cancer (HCC) therapy. In these systems either the Drug (D) and/or the NP <i>per se</i>, by exert biological activity. However, the high number of combinations of HCC cell lines, anti-cancer drugs, and/or NP to be tested slows down the assay process of new HCCNP systems. Nevertheless, the data scarcity and high complexity of these systems make difficult AI/ML studies. Here, a new dataset of HCCNP systems was compiled from literature. Herein 11 different Artificial Intelligence/Machine Learning (AI/ML) algorithms were used to seek the predictive models. The LDA and Random Forest (RF) models showed high values of sensitivity and specificity (&gt; 0.9) in training/validation series and 3-fold cross validation respectively. The new AI/ML models are able to predict 14 output properties (CC<sub>50 </sub>(µM), EC<sub>50 </sub>(µM), Inhibition (%), <i>etc</i>.) for all combinations of 54 different NP cores classes <i>vs</i>. 15 different coats and <i>vs</i>. 41 different cell lines allowing to short list the more interesting results for experimental assays. It may reduce the cost of the traditional trial and error procedures.