While advances in single-particle cryoEM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the so-called "preferred" orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep-learning-based software to address the preferred orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's capability of generating near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, β-galactosidases, and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred orientation problem.

中文翻译：

---

通过自监督深度学习克服冷冻电镜中的择优取向问题

虽然单粒子冷冻电镜的进步已经能够以原子分辨率测定大分子复合物的结构，但粒子取向偏差（所谓的“首选”取向问题）对于大多数样本来说仍然是一个复杂问题。现有的解决方案依赖于应用于样本的生化和物理策略，并且通常很复杂且具有挑战性。在这里，我们开发了 spIsoNet，这是一种基于端到端自监督深度学习的软件，用于解决首选方向问题。 spIsoNet 使用首选方向视图来恢复欠采样视图中的分子信息，从而提高了 3D 重建过程中的角度各向同性和粒子对准精度。我们展示了 spIsoNet 能够从具有有限视图的代表性生物系统（包括核糖体、β-半乳糖苷酶和以前难以处理的血凝素三聚体数据集）生成近各向同性重建的能力。 spIsoNet 还可以推广到改善子断层图平均中优先取向分子的图各向同性和粒子排列。因此，无需额外的样本制备程序，spIsoNet 即可为首选方向问题提供通用计算解决方案。