Box 3: HiC: a sequencing approach to characterize interactions over large genomic regions
HiC is an unbiased, high throughput method to detect chromatin-looping interactions between all loci in the genome144. HiC experiments have shown that genome structure and function are linked. In yeast, looping interaction groups form between highly transcribed genes, co-regulated genes (associated with different motifs) and gene ontology groups145, 146. Human HiC studies cluster looping interactions into three groups: self-interacting active gene–rich clusters, nonactive centromere-proximal clusters and nonactive centromere-distal clusters147. In bacteria, yeast and humans, chromatin loops place enhancers and promoters in close spatial proximity, which are thought to exist in topological domains bound by insulator proteins.
HiC libraries are created using proximity ligation. Tissue or cells are homogenized and crosslinked using 1% formaldehyde and cleaved using a restriction enzyme of choice. The 5′ overhangs are filled in with a biotinylated CTP residue and blunt-end ligation is performed under dilute conditions. The library is sheared for 300–500 bp size selection and biotinylated fragments are immunoprecipitated using streptavidin beads. The library is paired-end sequenced and aligned to the corresponding genome. Reads that map to two locations are scored as interactions and an interaction matrix from all interactions in the experiment is created. An expected interaction matrix is used to calculate the statistical significance of all scored experimental interactions.
Current HiC algorithms model looping interactions probabilistically by taking into account mappability, fragment length and GC percentage (which is correlated with gene density, banding patterns, repetitive content and chromatin marks)147. The best algorithm to date can resolve a human HiC map from cell lines to 5–10 kb using 63% mappable reads (300 million reads mapped from 500 million total reads) using four lanes of Illumina sequencing148. Problems from these algorithms arise from the probabilistic nature of proximity ligation. HiC detects the average chromatin structure within a population of cells. Failure to detect looping interactions does not mean they do not exist, but rather that the current method does not detect them147. In time, the efficiency of algorithms will further increase and the cost of sequencing will decrease, making HiC a more attractive approach for cell type–specific studies in brain.
Department of Pharmacology and Systems Therapeutics, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
- Ian Maze
Department of Neuroscience, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
- Ian Maze,
- Li Shen,
- Ningyi Shao,
- Amanda Mitchell,
- HaoSheng Sun,
- Schahram Akbarian &
- Eric J Nestler
Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
- Bin Zhang
Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.
- Benjamin A Garcia
Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York, USA.
- Amanda Mitchell &
- Schahram Akbarian
Laboratory of Chromatin Biology and Epigenetics, The Rockefeller University, New York, New York, USA.
- C David Allis
Competing financial interests
The authors declare no competing financial interests.
Benjamin A Garcia
C David Allis
Eric J Nestler