Maximum parsimony interpretation of chromatin capture experiments.

Area of Science:

• Genomics
• Computational Biology
• Molecular Biology

Background:

• Chromatin conformation capture techniques like Hi-C map genome structure by detecting physical contacts.
• Interpreting Hi-C signals as spatial proximity can lead to geometric inconsistencies, like violating the triangle inequality.
• These inconsistencies are often attributed to stochasticity or experimental noise.

Purpose of the Study:

• To present a novel graph-theoretic approach for characterizing the global geometric state of chromatin using Hi-C data.
• To demonstrate that geometric inconsistencies in Hi-C data can be explained by a mixture of cells in distinct conformational states.
• To identify properties of these postulated subpopulations and analyze their functional implications.

Main Methods:

• Development and implementation of a graph-theoretic approach to analyze Hi-C data.
• Identification of geometric conflicts within Hi-C datasets.
• Reconciliation of these conflicts by postulating homogeneous cell subpopulations.
• Analysis of functional annotations of differentially interacting genes between populations.

Main Results:

• Geometric conflicts in yeast Hi-C data can be explained by a small number of homogeneous cell populations.
• Four populations were sufficient to reconcile 95,000 impossible triangles, and eight explained 375,000.
• Inferred subpopulations show differential gene interactions, suggesting distinct transcriptional programs.

Conclusions:

• The observed geometric inconsistencies in Hi-C data are likely due to the presence of multiple, distinct cellular conformational states within a sample.
• The developed graph-theoretic method effectively identifies these subpopulations and their geometric properties.
• These findings suggest that different cell populations within a sample may engage in unique transcriptional activities.