Learning with multiple pairwise kernels for drug bioactivity prediction

• DOI: 10.1093/bioinformatics/bty277
• Publication Date: 2018-04-12T19:32:51Z
• Authors: Anna Cichonska; Tapio Pahikkala; Sandor Szedmak; Heli Julkunen; Antti Airola; Markus Heinonen; Tero Aittokallio; Juho Rousu
• Keywords: Kernel (algebra); Multiple kernel learning

Many inference problems in bioinformatics, including drug bioactivity prediction, can be formulated as pairwise learning problems, which one is interested making predictions for pairs of objects, e.g. drugs and their targets. Kernel-based approaches have emerged powerful tools solving that kind, especially multiple kernel (MKL) offers promising benefits it enables integrating various types complex biomedical information sources the form kernels, along with importance prediction task. However, immense size spaces remains a major bottleneck, existing MKL algorithms computationally infeasible even small number input pairs.