Transfer learning using attentions across atomic systems with graph neural networks (TAAG).

Recent advances in Graph Neural Networks (GNNs) have transformed the space of molecular and catalyst discovery. Despite the fact that the underlying physics across these domains remain the same, most prior work has focused on building domain-specific models either in small molecules or in materials. However, building large datasets across all domains is computationally expensive; therefore, the use of transfer learning (TL) to generalize to different domains is a promising but under-explored approach to this problem. To evaluate this hypothesis, we use a model that is pretrained on the Open Catalyst Dataset (OC20), and we study the model's behavior when fine-tuned for a set of different datasets and tasks. This includes MD17, the *CO adsorbate dataset, and OC20 across different tasks. Through extensive TL experiments, we demonstrate that the initial layers of GNNs learn a more basic representation that is consistent across domains, whereas the final layers learn more task-specific features. Moreover, these well-known strategies show significant improvement over the non-pretrained models for in-domain tasks with improvements of 53% and 17% for the *CO dataset and across the Open Catalyst Project (OCP) task, respectively. TL approaches result in up to 4× speedup in model training depending on the target data and task. However, these do not perform well for the MD17 dataset, resulting in worse performance than the non-pretrained model for few molecules. Based on these observations, we propose transfer learning using attentions across atomic systems with graph Neural Networks (TAAG), an attention-based approach that adapts to prioritize and transfer important features from the interaction layers of GNNs. The proposed method outperforms the best TL approach for out-of-domain datasets, such as MD17, and gives a mean improvement of 6% over a model trained from scratch.

Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences.

In the field of artificial intelligence, a combination of scale in data and model capacity enabled by unsupervised learning has led to major advances in representation learning and statistical generation. In the life sciences, the anticipated growth of sequencing promises unprecedented data on natural sequence diversity. Protein language modeling at the scale of evolution is a logical step toward predictive and generative artificial intelligence for biology. To this end, we use unsupervised learning to train a deep contextual language model on 86 billion amino acids across 250 million protein sequences spanning evolutionary diversity. The resulting model contains information about biological properties in its representations. The representations are learned from sequence data alone. The learned representation space has a multiscale organization reflecting structure from the level of biochemical properties of amino acids to remote homology of proteins. Information about secondary and tertiary structure is encoded in the representations and can be identified by linear projections. Representation learning produces features that generalize across a range of applications, enabling state-of-the-art supervised prediction of mutational effect and secondary structure and improving state-of-the-art features for long-range contact prediction.

Role of Urea-Aromatic Stacking Interactions in Stabilizing the Aromatic Residues of the Protein in Urea-Induced Denatured State.

A delicate balance of different types of intramolecular interactions makes the folded states of proteins marginally more stable than the unfolded states. Experiments use thermal, chemical, or mechanical stress to perturb the folding equilibrium for examining protein stability and the protein folding process. Elucidation of the mechanism by which chemical denaturants unfold proteins is crucial; this study explores the nature of urea-aromatic interactions relevant in urea-assisted protein denaturation. Free energy profiles corresponding to the unfolding of Trp-cage miniprotein in the presence and absence of urea at three different temperatures demonstrate the distortion of the hydrophobic core to be a crucial step. Exposure of the Trp6 residue to the solvent is found to be favored in the presence of urea. Previous experiments showed that urea has a high affinity for aromatic groups of proteins. We show here that this is due to the remarkable ability of urea to form stacking and NH-π interactions with aromatic groups of proteins. Urea-nucleobase stacking interactions have been shown to be crucial in urea-assisted RNA unfolding. Examination of these interactions using microsecond-long unrestrained simulations shows that urea-aromatic stacking interactions are stabilizing and long lasting. Further MD simulations, thermodynamic integration, and quantum mechanical calculations on aromatic model systems reveal that such interactions are possible for all the aromatic amino acid side-chains. Finally, we validate the ubiquitous nature of urea-aromatic stacking interactions by analyzing experimental structures of urea transporters and proteins crystallized in the presence of urea or urea derivatives.

Sumoylation of Sir2 differentially regulates transcriptional silencing in yeast.

Silent information regulator 2 (Sir2), the founding member of the conserved sirtuin family of NAD(+)-dependent histone deacetylase, regulates several physiological processes including genome stability, gene silencing, metabolism and life span in yeast. Within the nucleus, Sir2 is associated with telomere clusters in the nuclear periphery and rDNA in the nucleolus and regulates gene silencing at these genomic sites. How distribution of Sir2 between telomere and rDNA is regulated is not known. Here we show that Sir2 is sumoylated and this modification modulates the intra-nuclear distribution of Sir2. We identify Siz2 as the key SUMO ligase and show that multiple lysines in Sir2 are subject to this sumoylation activity. Mutating K215 alone counteracts the inhibitory effect of Siz2 on telomeric silencing. SUMO modification of Sir2 impairs interaction with Sir4 but not Net1 and, furthermore, SUMO modified Sir2 shows predominant nucleolar localization. Our findings demonstrate that sumoylation of Sir2 modulates distribution between telomeres and rDNA and this is likely to have implications for Sir2 function in other loci as well.

PARP-1 inhibition as a targeted strategy to treat Ewing's sarcoma.

Ewing's sarcoma family of tumors (ESFT) refers to aggressive malignancies which frequently harbor characteristic EWS-FLI1 or EWS-ERG genomic fusions. Here, we report that these fusion products interact with the DNA damage response protein and transcriptional coregulator PARP-1. ESFT cells, primary tumor xenografts, and tumor metastases were all highly sensitive to PARP1 inhibition. Addition of a PARP1 inhibitor to the second-line chemotherapeutic agent temozolamide resulted in complete responses of all treated tumors in an EWS-FLI1-driven mouse xenograft model of ESFT. Mechanistic investigations revealed that DNA damage induced by expression of EWS-FLI1 or EWS-ERG fusion genes was potentiated by PARP1 inhibition in ESFT cell lines. Notably, EWS-FLI1 fusion genes acted in a positive feedback loop to maintain the expression of PARP1, which was required for EWS-FLI-mediated transcription, thereby enforcing oncogene-dependent sensitivity to PARP-1 inhibition. Together, our findings offer a strong preclinical rationale to target the EWS-FLI1:PARP1 intersection as a therapeutic strategy to improve the treatment of ESFTs.