Fig. 6From: PS-MCL: parallel shotgun coarsened Markov clustering of protein interaction networksTime vs. average NCut for all datasets with b=1.5 for PS-MCL, MLR-MCL and MCL. We use 4 cores for PS-MCL. PS-MCL was 21% faster in running time and had 87% better NCut on average over all datasets compared with MLR-MCL. Although MCL is faster than PS-MCL for some networks, its average NCut is higher than that by PS-MCL for all networksBack to article page