A Von Mises-Fisher clustering modeling for the inference of orientational preferences of protein domains with scattering data

Abstract

This conference abstract describes a directional probabilistic model to infer the orientational preferences of protein domains from small‑angle X‑ray scattering (SAXS) data. Two beads connected by a thread represent a flexible protein linker; the relative position of the second bead follows magnetic orientation preferences. The proposed approach clusters precalculated bead models and uses a mixture of von Mises‑Fisher distributions to estimate orientation weights and select preferred orientations. The method aims to characterise multidomain proteins by inferring flexible linker orientations.

Publication
Joint Statistical Meetings 2021

Related