Table 3.

Approaches to account for systematic biases in Hi-C data

Approach	Model assumption*	Implementation‡	Computational speed	Refs
Yaffe and Tanay	Three systematic biases	Perl and R	Slow	77
HiCNorm	Three systematic biases	R	Fast	78
ICE	Equal visibility	Python	Fast	79
Knight and Ruiz	Equal visibility	JAVA	Fast	27
HiC-Pro	Equal visibility	Python and R	Very fast	80

ICE, iterative correction and eigenvector decomposition.

^*Model assumption refers to the inherent assumptions in the computational model used to account for bias in Hi-C data. These approaches can be classified based on their model assumptions: they are either explicit, assuming that systematic biases are known (three systematic biases), or implicit, assuming systemic biases are unknown and all the bias is captured by the sequencing coverage of each bin (equal visibility).

^‡Implementation refers to the programming language in which the normalization programme is written.