Characterizing the 3D structure and dynamics of chromosomes and proteins in a common contact matrix framework

Abstract

Conformational ensembles of biopolymers, whether proteins or chromosomes, can be described using contact matrices. Principal component analysis (PCA) on the contact data has been used to interrogate both protein and chromosome structures and/or dynamics. However, as these fields have developed separately, variants of PCA have emerged. Previously, a variant we hereby term Implicit-PCA (I-PCA) has been applied to chromosome contact matrices and revealed the spatial segregation of active and inactive chromatin. Separately, Explicit-PCA (E-PCA) has previously been applied to proteins and characterized their correlated structure fluctuations. Here, we swapped analysis methods (I-PCA and E-PCA), applying each to a different biopolymer type (chromosome or protein) than the one for which they were initially developed. We find that applying E-PCA to chromosome distance matrices derived from microscopy data can reveal the dominant motion (concerted fluctuation) of these chromosomes. Further, by applying E-PCA to Hi-C data across the human blood cell lineage, we isolated the aspects of chromosome structure that most strongly differentiate cell types. Conversely, when we applied I-PCA to simulation snapshots of proteins, the major component reported the consensus features of the structure, making this a promising approach for future analysis of semi-structured proteins.

Inferring chromosome radial organization from Hi-C data

“…In order to capture the major interaction trends between chromosomes from those distinct patterns, we applied principal component analysis (PCA) to these pairwise strong interaction matrices. PCA has been frequently applied to contacts within or between chromosomes to detect the spatial segregation of A/B compartments [ 29 , 60 – 62 ]. Further, the A/B compartment status of bins within a chromosome has been found to correlate with their lamina association and radial positioning, and therefore has been used in models predicting the lamin associations of domains within a chromosome [ 63 ].…”

3D Genome Structure Variation Across Cell Types Captured by Integrating Multi-omics

“…We next applied E-PCA to locate key sets of interactions that vary coordinately across cell types (step 2). The details of the E-PCA approach can be found in our previous study [12]. Briefly, E-PCA takes as an input a set of contact matrices (here, Hi-C data from different cell types), flattens each matrix into a vector of contacts, combines all these contact vectors into one matrix that contains every pairwise chromosome contact (rows) across all input cell types (columns), and then uses PCA to detect correlated patterns of contacts (PC1, PC2, etc.)…”

“…Specially, the conventional chromosome contact PCA highlights the spatial separation of active and inactive chromatin regions but is unable to dissect the major fluctuations of chromosome structure across different cell types. Our previous work applying E-PCA to microscopy data and Hi-C data demonstrated an alternative approach to capture dominant motions of chromosome structure [12]. Here, we apply this E-PCA approach to find major variations of genome structure for each chromosome across 35 Hi-C datasets from different cell types (Table 1).…”

“…To calculate the major variations in interaction patterns across Hi-C datasets, we applied the E-PCA method as previously described in the previous study [12]. We use the Spearman correlation matrices (where each entry is the correlation of contact patterns between the given pair of bins) rather than contact frequency matrices in order to emphasize the compartmentalization pattern rather than emphasizing only fluctuations in very local interactions, which may dominate the contact frequency matrix.…”

“…We previously reported the application of a contact matrix analysis originally developed for the study of a protein's structural dynamics, E-PCA, to a set of Hi-C matrices across nine different blood cell types on a portion of one chromosome [12]. Our results showed that the strength of compartment segregation was the major source of variation across those cell types, but we did not further investigate which particular A-A or B-B associations formed or separated coordinately.…”