Despite an explosion in the number of experimentally determined, atomically
detailed structures of biomolecules, many critical tasks in structural biology
remain data-limited. Whether performance in such tasks can be improved by using
large repositories of tangentially related structural data remains an open
question. To address this question, we focused on a central problem in biology:
predicting how proteins interact with one another---that is, which surfaces of
one protein bind to those of another protein. We built a training dataset, the
Database of Interacting Protein Structures (DIPS), that contains biases but is
two orders of magnitude larger than those used previously. We found that these
biases significantly degrade the performance of existing methods on
gold-standard data. Hypothesizing that assumptions baked into the hand-crafted
features on which these methods depend were the source of the problem, we
developed the first end-to-end learning model for protein interface prediction,
the Siamese Atomic Surfacelet Network (SASNet). Using only spatial coordinates
and identities of atoms, SASNet outperforms state-of-the-art methods trained on
gold-standard structural data, even when trained on only 3% of our new dataset.
Code and data available at this https URL

本研究基于一个大型数据库，通过开发学习模型 SASNet，改进了蛋白质相互作用预测的方法，证明了现有模型的性能不足以适用于大尺度数据，新模型展现出更好的性能。

3D 蛋白结构端到端学习用于界面预测

End-to-End Learning on 3D Protein Structure for Interface Prediction

High-throughput protein interaction detection methods are strongly affected
by false positive and false negative results. Focused experiments are needed to
complement the large-scale methods by validating previously detected
interactions but it is often difficult to decide which proteins to probe as
interaction partners. Developing reliable computational methods assisting this
decision process is a pressing need in bioinformatics. We show that we can use
the conserved properties of the protein network to identify and validate
interaction candidates. We apply a number of machine learning algorithms to the
protein connectivity information and achieve a surprisingly good overall
performance in predicting interacting proteins. Using a 'leave-one-out'
approach we find average success rates between 20-50% for predicting the
correct interaction partner of a protein. We demonstrate that the success of
these methods is based on the presence of conserved interaction motifs within
the network. A reference implementation and a table with candidate interacting
partners for each yeast protein are available at this http URL

通过机器学习算法使用蛋白质网络的保守性质确定并验证了蛋白质相互作用候选者，成功率在 20-50% 之间。

保守的网络模体允许蛋白质相互作用预测

Conserved network motifs allow protein-protein interaction prediction

The Saccharomyces cerevisiae protein-protein interaction map, as well as many
natural and man-made networks, shares the scale-free topology. The preferential
attachment model was suggested as a generic network evolution model that yields
this universal topology. However, it is not clear that the model assumptions
hold for the protein interaction network. Using a cross genome comparison we
show that (a) the older a protein, the better connected it is, and (b) The
number of interactions a protein gains during its evolution is proportional to
its connectivity. Therefore, preferential attachment governs the protein
network evolution. The evolutionary mechanism leading to such preference and
some implications are discussed.

通过跨基因组比较，我们表明优先附着模型主宰蛋白质网络的进化，并讨论了导致这种偏好的进化机制及其一些影响。