Computational Discovery of TTF Molecules with Deep Generative Models

We present a computational workflow based on quantum chemical calculations and generative models based on deep neural networks for the discovery of novel materials. We apply the developed workflow to search for molecules suitable for the fusion of triplet-triplet excitations (triplet-triplet fusion, TTF) in blue OLED devices. By applying generative machine learning models, we have been able to pinpoint the most promising regions of the chemical space for further exploration. Another neural network based on graph convolutions was trained to predict excitation energies; with this network, we estimate the alignment of energy levels and filter molecules before running time-consuming quantum chemical calculations. We present a comprehensive computational evaluation of several generative models, choosing a modification of the Junction Tree VAE (JT-VAE) as the best one in this application. The proposed approach can be useful for computer-aided design of materials with energy level alignment favorable for efficient energy transfer, triplet harvesting, and exciton fusion processes, which are crucial for the development of the next generation OLED materials.

Building high accuracy emulators for scientific simulations with deep neural architecture search

Computer simulations are invaluable tools for scientific discovery. However, accurate simulations are often slow to execute, which limits their applicability to extensive parameter exploration, large-scale data analysis, and uncertainty quantification. A promising route to accelerate simulations by building fast emulators with machine learning requires large training datasets, which can be prohibitively expensive to obtain with slow simulations. Here we present a method based on neural architecture search to build accurate emulators even with a limited number of training data. The method successfully emulates simulations in 10 scientific cases including astrophysics, climate sci-ence, biogeochemistry, high energy density physics, fusion energy, and seismology, using the same super-architecture, algorithm, and hyperparameters. Our approach also inherently provides emulator uncertainty estimation, adding further confidence in their use. We anticipate this work will accelerate research involving expensive simulations, allow more extensive parameters exploration, and enable new, previously unfeasible computational discovery.

Computational Discovery of a miRNA and its Putative Target Genes in Ziziphus Jujuba using Genome-wide Expressed Sequence Tags

Background Ziziphus jujuba is an important fruit crop which is increasingly becoming popular among consumers due to its medicinal properties. Increasing worldwide demand for the fruit poses new challenges to the industry which includes the need for accelerated cultivar development of jujubes. To embark on cultivar development with improved traits such as high yield and disease resistance, molecular and conventional breeding, and genetic engineering become imperative. But inadequate trait-enhancing alleles or gene pleiotropism limit the direct use of several identified genes. To overcome these issues, microRNAs (miRNAs) can be utilized in breeding of jujubes as genetic modulators to fine-tune the regulation of gene expression, thus the discovery of miRNAs becomes important. Methods and results In this study using a computational approach, we identified one potential miRNA (zju-miR-215-3p) from 2904 expressed sequence tags. The miRNA showed down regulation of five target proteins (AP-2 complex subunit alpha, C2H2-type domain-containing protein, sentrin-specific protease 1, hydrolase_4 domain-containing protein and putative alpha-ketoglutarate-dependent dioxygenase) and their suppression appears to be helpful to the plant to overcome stress conditions. Conclusion The miRNA identified in this study is associated with five potential target proteins, most of which are implicated in metabolic and developmental processes associated with plant growth and reproduction. Future studies are necessary to validate the miRNA by RNA sequencing and to confirm the molecular functions of the down regulations of target proteins.

Large-scale discovery of recombinases for integrating DNA into the human genome

Recent microbial genome sequencing efforts have revealed a vast reservoir of mobile genetic elements containing integrases that could be useful genome engineering tools. Large serine recombinases (LSRs), such as Bxb1 and PhiC31, are bacteriophage-encoded integrases that can facilitate the insertion of phage DNA into bacterial genomes. However, only a few LSRs have been previously characterized and they have limited efficiency in human cells. Here, we developed a systematic computational discovery workflow that searches across the bacterial tree of life to expand the diversity of known LSRs and their cognate DNA attachment sites by >100-fold. We validated this approach via experimental characterization of LSRs, leading to three classes of LSRs distinguished from one another by their efficiency and specificity. We identify landing pad LSRs that efficiently integrate into native attachment sites in a human cell context, human genome-targeting LSRs with computationally predictable pseudosites, and multi-targeting LSRs that can unidirectionally integrate cargos with similar efficiency and superior specificity to commonly used transposases. LSRs from each category were functionally characterized in human cells, overall achieving up to 7-fold higher plasmid recombination than Bxb1 and genome insertion efficiencies of 40-70% with cargo sizes over 7 kb. Overall, we establish a paradigm for the large-scale discovery of microbial recombinases directly from sequencing data and the reconstruction of their target sites. This strategy provided a rich resource of over 60 experimentally characterized LSRs that can function in human cells and thousands of additional candidates for large-payload genome editing without double-stranded DNA breaks.

Formation of a single quasicrystal upon collision of multiple grains

Quasicrystals exhibit long-range order but lack translational symmetry. When grown as single crystals, they possess distinctive and unusual properties owing to the absence of grain boundaries. Unfortunately, conventional methods such as bulk crystal growth or thin film deposition only allow us to synthesize either polycrystalline quasicrystals or quasicrystals that are at most a few centimeters in size. Here, we reveal through real-time and 3D imaging the formation of a single decagonal quasicrystal arising from a hard collision between multiple growing quasicrystals in an Al-Co-Ni liquid. Through corresponding molecular dynamics simulations, we examine the underlying kinetics of quasicrystal coalescence and investigate the effects of initial misorientation between the growing quasicrystalline grains on the formation of grain boundaries. At small misorientation, coalescence occurs following rigid rotation that is facilitated by phasons. Our joint experimental-computational discovery paves the way toward fabrication of single, large-scale quasicrystals for novel applications.

PPIDomainMiner: Inferring domain-domain interactions from multiple sources of protein-protein interactions

Many biological processes are mediated by protein-protein interactions (PPIs). Because protein domains are the building blocks of proteins, PPIs likely rely on domain-domain interactions (DDIs). Several attempts exist to infer DDIs from PPI networks but the produced datasets are heterogeneous and sometimes not accessible, while the PPI interactome data keeps growing. We describe a new computational approach called “PPIDM” (Protein-Protein Interactions Domain Miner) for inferring DDIs using multiple sources of PPIs. The approach is an extension of our previously described “CODAC” (Computational Discovery of Direct Associations using Common neighbors) method for inferring new edges in a tripartite graph. The PPIDM method has been applied to seven widely used PPI resources, using as “Gold-Standard” a set of DDIs extracted from 3D structural databases. Overall, PPIDM has produced a dataset of 84, 552 non-redundant DDIs. Statistical significance (p-value) is calculated for each source of PPI and used to classify the PPIDM DDIs in Gold (9, 175 DDIs), Silver (24, 934 DDIs) and Bronze (50, 443 DDIs) categories. Dataset comparison reveals that PPIDM has inferred from the 2017 releases of PPI sources about 46% of the DDIs present in the 2020 release of the 3did database, not counting the DDIs present in the Gold-Standard. The PPIDM dataset contains 10, 229 DDIs that are consistent with more than 13, 300 PPIs extracted from the IMEx database, and nearly 23, 300 DDIs (27.5%) that are consistent with more than 214, 000 human PPIs extracted from the STRING database. Examples of newly inferred DDIs covering more than 10 PPIs in the IMEx database are provided. Further exploitation of the PPIDM DDI reservoir includes the inventory of possible partners of a protein of interest and characterization of protein interactions at the domain level in combination with other methods. The result is publicly available at http://ppidm.loria.fr/.