Introduction

In MCQD-ML algorithm, we have extended the MCQD algorithm with an initial phase of machine learning-based prediction of T limit parameter that is best suited for each input graph. We show empirically that the resulting new algorithm, MCQD-ML improves search speeds on certain types of graphs, most notably for molecular docking graphs that are used in drug design where they determine energetically favorable conformations of small molecules in a protein binding site. In such cases, the speedup is 2-fold.

Download

The MCQD-ML algorithm source code is freely available at GitHub.

Graphs

Small protein product graphs, full protein product graphs and docking graphs that were used in the paper are available:

• Small protein product graphs (see Table 4 in Reba et al. paper)
• Full protein product graphs (see Table 5 in Reba et al. paper)
• Docking graphs (see Table 6 in Reba et al. paper)

Please Cite

• Kristjan Reba, Matej Guid, Kati Rozman, Dušanka Janežič, Janez Konc. Exact Maximum Clique Algorithm for Different Graph Types Using Machine Learning. Mathematics, 2022, 10, 97. Open Access