Evaluating the use of absolute binding free energy in the fragment optimization process.

Key to the fragment optimization process is the need to accurately capture the changes in affinity that are associated with a given set of chemical modifications. Due to the weakly binding nature of fragments, this has proven to be a challenging task, despite recent advancements in leveraging experimental and computational methods. In this work, we evaluate the use of Absolute Binding Free Energy (ABFE) calculations in guiding fragment optimization decisions, retrospectively calculating binding free energies for 59 ligands across 4 fragment elaboration campaigns. We first demonstrate that ABFEs can be used to accurately rank fragment-sized binders with an overall Spearman’s r of 0.89 and a Kendall τ of 0.67, although often deviating from experiment in absolute free energy values with an RMSE of 2.75 kcal/mol. We then also show that in several cases, retrospective fragment optimization decisions can be supported by the ABFE calculations. Cases that were not supported were often limited by large uncertainties in the free energy estimates, however generally the right direction in ΔΔG is still observed. Comparing against cheaper endpoint methods, namely Nwat-MM/GBSA, we find that ABFEs offer better outcomes in ranking binders, improving correlation metrics, although a similar confidence in retrospective synthetic decisions is achieved. Our results indicate that ABFE calculations are currently at the level of accuracy that can be usefully employed to gauge which fragment elaborations are likely to offer the best gains in affinity.

TLCrys: Transfer Learning Based Method for Protein Crystallization Prediction

X-ray diffraction technique is one of the most common methods of ascertaining protein structures, yet only 2–10% of proteins can produce diffraction-quality crystals. Several computational methods have been proposed so far to predict protein crystallization. Nevertheless, the current state-of-the-art computational methods are limited by the scarcity of experimental data. Thus, the prediction accuracy of existing models hasn’t reached the ideal level. To address the problems above, we propose a novel transfer-learning-based framework for protein crystallization prediction, named TLCrys. The framework proceeds in two steps: pre-training and fine-tuning. The pre-training step adopts attention mechanism to extract both global and local information of the protein sequences. The representation learned from the pre-training step is regarded as knowledge to be transferred and fine-tuned to enhance the performance of crystalization prediction. During pre-training, TLCrys adopts a multi-task learning method, which not only improves the learning ability of protein encoding, but also enhances the robustness and generalization of protein representation. The multi-head self-attention layer guarantees that different levels of the protein representation can be extracted by the fine-tuned step. During transfer learning, the fine-tuning strategy used by TLCrys improves the task-specialized learning ability of the network. Our method outperforms all previous predictors significantly in five crystallization stages of prediction. Furthermore, the proposed methodology can be well generalized to other protein sequence classification tasks.

A comparison on predicting functional impact of genomic variants

ABSTRACT Single-nucleotide polymorphism (SNPs) may cause the diverse functional impact on RNA or protein changing genotype and phenotype, which may lead to common or complex diseases like cancers. Accurate prediction of the functional impact of SNPs is crucial to discover the ‘influential’ (deleterious, pathogenic, disease-causing, and predisposing) variants from massive background polymorphisms in the human genome. Increasing computational methods have been developed to predict the functional impact of variants. However, predictive performances of these computational methods on massive genomic variants are still unclear. In this regard, we systematically evaluated 14 important computational methods including specific methods for one type of variant and general methods for multiple types of variants from several aspects; none of these methods achieved excellent (AUC ≥ 0.9) performance in both data sets. CADD and REVEL achieved excellent performance on multiple types of variants and missense variants, respectively. This comparison aims to assist researchers and clinicians to select appropriate methods or develop better predictive methods.

Exploring Ancient Networks

A small archive of texts from ancient Iraq is used to demonstrate an approach to network analysis in which traditional close reading and computational text analysis go hand-in-hand. The computational methods produce tables and graphs that link back to online editions of the primary material, enabling the user to check the results.