Shadman, H. | Search bioRxiv

bioRxiv2025

Resolving Local and Global Conformational Heterogeneity of the Human Intrinsically Disordered Proteome

Linking the sequences of intrinsically disordered regions (IDRs) to their structural ensembles and biological functions remains a central challenge in understanding how disorder encodes activity. A recent study has shown that human IDRs with different levels of compactness, as measured by Florys exponent, are associated with specific cellular functions and localizations, and Florys exponent can be predicted from sequence features with reasonable accuracy. However, IDRs are known to sample highly heterogenous conformations that can be masked by ensemble-averaged metrics such as Florys exponent. Here, we introduce a simple framework that resolves heterogeneous conformations. We pair two polymer physics descriptors, shape ratio (Rs) and relative shape anisotropy (RSA), to construct joint two-dimensional (RSA, Rs) maps at both the global and local (subchain) scales. We show that sequences with similar Florys exponent can display strikingly different two-dimensional conformational maps that reflect differences in charge patterning. IDRs with similar global maps can exhibit markedly different local maps that can be linked to local sequence variations. This leads to the identification of a class of IDRs that appear non-compact globally but contain locally compact subchains. These locally compact IDRs are found to be associated with similar GO functional and cellular localization enrichments and phase-separation propensities as globally compact IDRs. Our framework moves beyond ensemble-averaged descriptors, providing new tools that capture the intrinsic heterogeneity of IDR conformations and thus offer new opportunities to link IDR sequences with functions. Significance StatementIntrinsically disordered proteins (IDPs) and regions (IDRs) exhibit highly heterogeneous conformational ensembles, yet studies routinely use ensemble-averaged metrics to characterize them that obscure this heterogeneity. We present a simple, general framework that quantifies structural heterogeneity in human IDRs. This resolves forms of structural heterogeneity that are otherwise overlooked and enhances our understanding of how IDR structures relate to their sequences and functions. A python package for immediate implementation of this framework is provided.

biochemistry↗

bioRxiv2024

A Machine Learning-Based Investigation of Integrin Expression Patterns in Cancer and Metastasis

BackgroundIntegrins, a family of transmembrane receptor proteins, play complex roles in cancer development and metastasis. These roles could be better delineated through machine learning of transcriptomic data to reveal relationships between integrin expression patterns and cancer. MethodsWe collected publicly available RNA-Seq integrin expression from 8 healthy tissues and their corresponding tumors, along with data from metastatic breast cancer. We then used machine learning methods, including t-SNE visualization and Random Forest classification, to investigate changes in integrin expression patterns. ResultsIntegrin expression varied across tissues and cancers, and between healthy and cancer samples from the same tissue, enabling the creation of models that classify samples by tissue or disease status. The integrins whose expression was important to these classifiers were identified. For example, ITGA7 was key to classification of breast samples by disease status. Analysis in breast tissue revealed that cancer rewires co-expression for most integrins, but the co-expression relationships of some integrins remain unchanged in healthy and cancer samples. Integrin expression in primary breast tumors differed from their metastases, with liver metastasis notably having reduced expression. ConclusionsIntegrin expression patterns vary widely across tissues and are greatly impacted by cancer. Machine learning of these patterns can effectively distinguish samples by tissue or disease status.

cancer biology↗