Hi-C vs Micro-C: 3D Genome Methods Compared

Updated 7 November 2026 · Standard formula, shown below

Hi-C and Micro-C are both chromosome conformation capture techniques that map the 3D organization of the genome, but they differ in resolution and the types of interactions they detect. Hi-C uses restriction enzymes to cut DNA, capturing interactions at a resolution of 1-10 kb, while Micro-C uses a non-specific nuclease to digest DNA into much smaller fragments, achieving nucleosome-level resolution (~200 bp). This article explains how both methods work, their key differences, and how to choose between them for your research.

Background: The 3D Genome and Chromosome Conformation Capture

The human genome is about 2 meters of DNA packed into a nucleus only 10 micrometers across. This packaging is not random: the genome folds into a hierarchical 3D structure that brings distant regulatory elements (like enhancers) into physical contact with their target genes. This spatial organization is critical for gene regulation, DNA replication, and repair.

Chromosome conformation capture (3C) techniques were developed to measure how often any two genomic regions physically interact. The basic principle is:

Crosslink DNA-bound proteins to DNA using formaldehyde, freezing the 3D structure.
Cut the DNA with an enzyme, then ligate the cut ends together in a way that only happens if they were in close spatial proximity.
Sequence the ligated junctions to identify which regions interacted.

Hi-C is the genome-wide version of this approach, using a restriction enzyme (typically HindIII or MboI) that cuts at specific 6-base-pair recognition sites. This creates fragments averaging 1-4 kb in size, limiting resolution to that range. Micro-C, developed later, replaces the restriction enzyme with a non-specific nuclease (MNase) that cuts DNA essentially everywhere, producing fragments of ~200 bp. This allows detection of interactions at the scale of single nucleosomes.

How Hi-C Works: Restriction Enzyme-Based Capture

Hi-C is the most widely used 3C method. The protocol involves:

Crosslinking: Cells are treated with formaldehyde to covalently link proteins to DNA, preserving spatial contacts.
Digestion: The crosslinked DNA is cut with a restriction enzyme, such as HindIII (recognizes AAGCTT) or MboI (GATC). This produces sticky ends.
Fill-in and label: The sticky ends are filled in with biotin-labeled nucleotides, tagging each fragment end.
Ligation: Under dilute conditions, DNA ends that are in close proximity are ligated together, creating chimeric molecules.
Purification and sequencing: DNA is sheared, biotin-labeled junctions are pulled down, and paired-end sequencing identifies which two genomic regions were joined.

The resolution of Hi-C is determined by the frequency of restriction sites. For example, HindIII cuts on average every 4 kb, so the smallest fragments are around 1 kb, and interactions at finer scales are missed. Typical Hi-C experiments achieve 1-10 kb resolution, which is sufficient to identify topologically associating domains (TADs) and loops, but not individual nucleosome contacts.

How Micro-C Works: Nucleosome-Level Resolution

Micro-C was introduced in 2017 to overcome Hi-C’s resolution limit. The key change is the digestion step:

Crosslinking: Same as Hi-C, using formaldehyde.
Digestion with MNase: Micrococcal nuclease (MNase) is a non-specific endonuclease that cuts DNA between nucleosomes, producing mononucleosome-sized fragments (~147 bp of DNA wrapped around a histone octamer, plus linker).
Ligation: The blunt-ended fragments are ligated together under dilute conditions.
Sequencing: Paired-end sequencing identifies the ligated junctions.

Because MNase cuts everywhere, the resolution is limited only by the fragment size, which is about 200 bp. This allows Micro-C to detect interactions between individual nucleosomes, revealing fine-scale structures such as nucleosome phasing, promoter-enhancer contacts, and even the internal structure of TADs. Micro-C also tends to have a higher signal-to-noise ratio because it captures more ligation events per unit of DNA.

Key Differences: Resolution, Cost, and Data Analysis

The main differences between Hi-C and Micro-C are:

Resolution: Hi-C: 1-10 kb; Micro-C: ~200 bp (nucleosome level).
Enzyme: Hi-C uses restriction enzymes (HindIII, MboI); Micro-C uses MNase.
Fragment size: Hi-C fragments are 1-4 kb; Micro-C fragments are ~200 bp.
Cost: Micro-C is generally more expensive due to the need for more sequencing depth (to achieve the same coverage at higher resolution) and the MNase enzyme.
Data analysis: Both use similar pipelines (alignment, filtering, binning), but Micro-C data requires higher-resolution binning (e.g., 500 bp vs 10 kb) and more memory.

In practice, Hi-C is sufficient for many studies, such as identifying TADs and large loops. Micro-C is preferred when you need to see details like enhancer-promoter interactions at base-pair resolution, or when studying chromatin structure in small genomes like yeast.

Choosing Between Hi-C and Micro-C

Your choice depends on your biological question and budget:

Use Hi-C if: You are studying large-scale genome organization (TADs, compartments, loops) in mammalian genomes; you have limited sequencing budget; you need to compare with existing Hi-C datasets.
Use Micro-C if: You need to resolve interactions at the nucleosome level, such as promoter-enhancer contacts, or if you are studying organisms with small genomes (yeast, bacteria) where higher resolution is feasible.
Consider hybrid approaches: Some studies use both, or use Hi-C for initial discovery and Micro-C for fine mapping of specific regions.

For example, a study of gene regulation in human cells might use Hi-C to find TADs, then Micro-C to zoom in on a 50 kb region to see exactly which nucleosomes contact the promoter.

Common Problems and Troubleshooting

Both methods have common pitfalls:

Low crosslinking efficiency: Leads to few captured interactions. Fix: increase formaldehyde concentration or time (but avoid over-crosslinking, which can reduce signal).
Incomplete digestion: Results in large fragments and poor resolution. Fix: optimize enzyme concentration and digestion time.
Ligation artifacts: Random ligation can create false contacts. Fix: use proper dilution conditions and include biological replicates.
Biased sequencing: Some regions amplify more than others. Fix: use appropriate library preparation and bioinformatics normalization (e.g., ICE).

For Micro-C, a specific issue is MNase over-digestion, which can degrade DNA. It is important to titrate the MNase concentration and check fragment sizes on a gel before proceeding.

Alternative Methods for Studying 3D Genome Structure

Hi-C and Micro-C are not the only options. Other techniques include:

3C (one-to-one): Measures interaction between two specific loci using PCR. Low throughput but cheap.
4C (one-to-all): Captures interactions between one viewpoint and all other loci. Useful for studying a specific region.
5C (many-to-many): Uses multiplexed ligation to capture interactions among many loci in a region.
ChIA-PET and HiChIP: Combine chromatin immunoprecipitation (ChIP) with Hi-C to enrich for interactions involving a specific protein (e.g., CTCF or RNA Pol II).
GAM (genome architecture mapping): Uses cryosectioning and DNA sequencing to map contacts without ligation, avoiding some biases.
SPRITE: Uses split-pool recognition of interactions, allowing detection of multi-way contacts.

Each method has trade-offs in resolution, throughput, cost, and bias. For most researchers, Hi-C remains the default, with Micro-C as a higher-resolution upgrade when needed.

Frequently Asked Questions

Can Micro-C replace Hi-C entirely?

Not yet. Micro-C offers higher resolution, but it is more expensive and requires more sequencing depth. For many applications, Hi-C data is sufficient, and there is a vast amount of public Hi-C data for comparison. Micro-C is best used when nucleosome-level resolution is essential.

What is the minimum sequencing depth for Micro-C?

For a mammalian genome, a typical Hi-C experiment uses 100-500 million read pairs. Micro-C, because it captures more fragments, may require 500 million to 1 billion read pairs to achieve comparable coverage at higher resolution. For smaller genomes like yeast, 10-50 million reads may suffice.

Does Micro-C work with single cells?

Single-cell Hi-C is possible, and single-cell Micro-C has been demonstrated in a few studies, but it is technically challenging due to the low amount of starting material. The MNase digestion step is harder to control in single cells, and the efficiency is lower.

How do I analyze Micro-C data?

Most Hi-C analysis tools (e.g., HiC-Pro, Juicer, HiCExplorer) can process Micro-C data, but you need to adjust the bin size to 500 bp or 1 kb instead of 10 kb. Some tools have specific parameters for Micro-C, such as the “micro-c” mode in Juicer.

Is Micro-C better for detecting enhancer-promoter interactions?

Yes, because enhancer-promoter contacts often occur at sub-kilobase scales. Micro-C can resolve these interactions more precisely than Hi-C, reducing false positives and improving the identification of causal regulatory contacts.

What is the difference between Micro-C and HiChIP?

HiChIP combines Hi-C with antibody-based enrichment for a specific protein, so it only shows interactions involving that protein. Micro-C shows all interactions at high resolution. If you want to study a specific protein’s role in 3D structure, HiChIP is more targeted; if you want a global high-resolution map, Micro-C is better.

Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.