FISHnet: detecting chromatin domains in single-cell sequential Oligopaints imaging data.

Area of Science:

• Genomics
• Computational Biology
• Molecular Biology

Background:

• Sequential Oligopaints DNA FISH enables high-resolution analysis of higher-order genome folding.
• Existing algorithms are insufficient for analyzing complex 3D genome features in this type of data.

Purpose of the Study:

• To introduce FISHnet, a novel graph theory-based algorithm for detecting 3D genome structures in sequential Oligopaints data.
• To provide a robust method for analyzing single-allele genome folding patterns.

Main Methods:

• Developed FISHnet, a graph theory method optimizing network modularity.
• Applied FISHnet to simulated and real single-allele imaging data.
• Utilized statistical tests to identify cell-type-specific folding patterns.

Main Results:

• FISHnet accurately and sensitively detects chromatin domains and boundaries in imaging data.
• Confirmed the presence of nested domains (TADs, subTADs) at the single-allele level, not just in ensemble data.
• Demonstrated FISHnet's capability across multiple published Oligopaints datasets.

Conclusions:

• FISHnet is an effective tool for analyzing 3D genome organization from single-allele imaging data.
• The findings suggest that TADs and subTADs are observable at the single-allele level.
• The freely available FISHnet code will facilitate future research into genome function and folding variation.