Search

In this paper, a new Information Foraging approach based on Elephant Herding Optimization (EHO) is proposed and tested on social media. We adapted the original EHO algorithm and combined it with the information foraging theory. In order to test our approach, we constructed a dataset containing more than one million tweets collected during the second semester of 2020. The results are very satisfying and show the ability of our approach to improve the information foraging process both in terms of relevance and response time. To further evaluate our system, we held a comparative study with another well-known metaheuristic applied to information foraging, namely Ant Colony Optimization. The outcomes show the superiority of our proposal.

Hypoxia is a key characteristic of the tumor microenvironment, too rarely considered during drug development due to the lack of a user-friendly method to culture naturally hypoxic 3D tumor models. In this study, we used soft lithography to engineer a microfluidic platform allowing the culture of up to 240 naturally hypoxic tumor spheroids within an 80 mm by 82.5 mm chip. These jumbo spheroids on a chip are the largest to date (>750 µm), and express gold-standard hypoxic protein CAIX at their core only, a feature absent from smaller spheroids of the same cell lines. Using histopathology, we investigated response to combined radiotherapy (RT) and hypoxic prodrug Tirapazamine (TPZ) on our jumbo spheroids produced using two sarcoma cell lines (STS117 and SK-LMS-1). Our results demonstrate that TPZ preferentially targets the hypoxic core (STS117: p = 0.0009; SK-LMS-1: p = 0.0038), but the spheroids’ hypoxic core harbored as much DNA damage 24 h after irradiation as normoxic spheroid cells. These results validate our microfluidic device and jumbo spheroids as potent fundamental and pre-clinical tools for the study of hypoxia and its effects on treatment response.