Planet-compatible pathways for transitioning the chemical industry.

Chemical products, such as plastics, solvents, and fertilizers, are essential for supporting modern lifestyles. Yet, producing, using, and disposing of chemicals creates adverse environmental impacts which threaten the industry's license to operate. This study presents seven planet-compatible pathways toward 2050 employing demand-side and supply-side interventions with cumulative total investment costs of US$1.2-3.7 trillion. Resource efficiency and circularity interventions reduce global chemicals demand by 23 to 33% and are critical for mitigating risks associated with using fossil feedstocks and carbon capture and sequestration, and constraints on available biogenic and recyclate feedstocks. Replacing fossil feedstocks with biogenic/air-capture sources, shifting carbon destinations from the atmosphere to ground, and electrifying/decarbonizing energy supply for production technologies could enable net negative emissions of 0.5 GtCO2eq y-1 across non-ammonia chemicals, while still delivering essential chemical-based services to society.

In vitro import experiments with semi-intact cells suggest a role of the Sec61 paralog Ssh1 in mitochondrial biogenesis.

Mitochondrial biogenesis relies on the synthesis of hundreds of different precursor proteins in the cytosol and their subsequent import into the organelle. Recent studies suggest that the surface of the endoplasmic reticulum (ER) actively contributes to the targeting of some mitochondrial precursors. In the past, in vitro import experiments with isolated mitochondria proved to be extremely powerful to elucidate the individual reactions of the mitochondrial import machinery. However, this in vitro approach is not well suited to study the influence of non-mitochondrial membranes. In this study, we describe an in vitro system using semi-intact yeast cells to test a potential import relevance of the ER proteins Erg3, Lcb5 and Ssh1, all being required for efficient mitochondrial respiration. We optimized the conditions of this experimental test system and found that cells lacking Ssh1, a paralog of the Sec61 translocation pore, show a reduced import efficiency of mitochondrial precursor proteins. Our results suggest that Ssh1, directly or indirectly, increases the efficiency of the biogenesis of mitochondrial proteins. Our findings are compatible with a functional interdependence of the mitochondrial and the ER protein translocation systems.

Import of ribosomal proteins into yeast mitochondria.

Mitochondrial ribosomes of baker's yeast contain at least 78 protein subunits. All but one of these proteins are nuclear-encoded, synthesized on cytosolic ribosomes, and imported into the matrix for biogenesis. The import of matrix proteins typically relies on N-terminal mitochondrial targeting sequences that form positively charged amphipathic helices. Interestingly, the N-terminal regions of many ribosomal proteins do not closely match the characteristics of matrix targeting sequences, suggesting that the import processes of these proteins might deviate to some extent from the general import route. So far, the biogenesis of only two ribosomal proteins, Mrpl32 and Mrp10, was studied experimentally and indeed showed surprising differences to the import of other preproteins. In this review article we summarize the current knowledge on the transport of proteins into the mitochondrial matrix, and thereby specifically focus on proteins of the mitochondrial ribosome.

Single-nucleoid architecture reveals heterogeneous packaging of mitochondrial DNA.

Cellular metabolism relies on the regulation and maintenance of mitochondrial DNA (mtDNA). Hundreds to thousands of copies of mtDNA exist in each cell, yet because mitochondria lack histones or other machinery important for nuclear genome compaction, it remains unresolved how mtDNA is packaged into individual nucleoids. In this study, we used long-read single-molecule accessibility mapping to measure the compaction of individual full-length mtDNA molecules at near single-nucleotide resolution. We found that, unlike the nuclear genome, human mtDNA largely undergoes all-or-none global compaction, with most nucleoids existing in an inaccessible, inactive state. Highly accessible mitochondrial nucleoids are co-occupied by transcription and replication components and selectively form a triple-stranded displacement loop structure. In addition, we showed that the primary nucleoid-associated protein TFAM directly modulates the fraction of inaccessible nucleoids both in vivo and in vitro, acting consistently with a nucleation-and-spreading mechanism to coat and compact mitochondrial nucleoids. Together, these findings reveal the primary architecture of mtDNA packaging and regulation in human cells.

Orbital-free bond breaking via machine learning.

Using a one-dimensional model, we explore the ability of machine learning to approximate the non-interacting kinetic energy density functional of diatomics. This nonlinear interpolation between Kohn-Sham reference calculations can (i) accurately dissociate a diatomic, (ii) be systematically improved with increased reference data and (iii) generate accurate self-consistent densities via a projection method that avoids directions with no data. With relatively few densities, the error due to the interpolation is smaller than typical errors in standard exchange-correlation functionals.

The unfolded protein response of the endoplasmic reticulum supports mitochondrial biogenesis by buffering nonimported proteins.

Almost all mitochondrial proteins are synthesized in the cytosol and subsequently targeted to mitochondria. The accumulation of nonimported precursor proteins occurring upon mitochondrial dysfunction can challenge cellular protein homeostasis. Here we show that blocking protein translocation into mitochondria results in the accumulation of mitochondrial membrane proteins at the endoplasmic reticulum, thereby triggering the unfolded protein response (UPRER). Moreover, we find that mitochondrial membrane proteins are also routed to the ER under physiological conditions. The level of ER-resident mitochondrial precursors is enhanced by import defects as well as metabolic stimuli that increase the expression of mitochondrial proteins. Under such conditions, the UPRER is crucial to maintain protein homeostasis and cellular fitness. We propose the ER serves as a physiological buffer zone for those mitochondrial precursors that cannot be immediately imported into mitochondria while engaging the UPRER to adjust the ER proteostasis capacity to the extent of precursor accumulation.

Finding density functionals with machine learning.

Machine learning is used to approximate density functionals. For the model problem of the kinetic energy of noninteracting fermions in 1D, mean absolute errors below 1 kcal/mol on test densities similar to the training set are reached with fewer than 100 training densities. A predictor identifies if a test density is within the interpolation region. Via principal component analysis, a projected functional derivative finds highly accurate self-consistent densities. The challenges for application of our method to real electronic structure problems are discussed.

Multi-task learning for pKa prediction.

Many compound properties depend directly on the dissociation constants of its acidic and basic groups. Significant effort has been invested in computational models to predict these constants. For linear regression models, compounds are often divided into chemically motivated classes, with a separate model for each class. However, sometimes too few measurements are available for a class to build a reasonable model, e.g., when investigating a new compound series. If data for related classes are available, we show that multi-task learning can be used to improve predictions by utilizing data from these other classes. We investigate performance of linear Gaussian process regression models (single task, pooling, and multi-task models) in the low sample size regime, using a published data set (n = 698, mostly monoprotic, in aqueous solution) divided beforehand into 15 classes. A multi-task regression model using the intrinsic model of co-regionalization and incomplete Cholesky decomposition performed best in 85% of all experiments. The presented approach can be applied to estimate other molecular properties where few measurements are available.

Optimizing transition states via kernel-based machine learning.

We present a method for optimizing transition state theory dividing surfaces with support vector machines. The resulting dividing surfaces require no a priori information or intuition about reaction mechanisms. To generate optimal dividing surfaces, we apply a cycle of machine-learning and refinement of the surface by molecular dynamics sampling. We demonstrate that the machine-learned surfaces contain the relevant low-energy saddle points. The mechanisms of reactions may be extracted from the machine-learned surfaces in order to identify unexpected chemically relevant processes. Furthermore, we show that the machine-learned surfaces significantly increase the transmission coefficient for an adatom exchange involving many coupled degrees of freedom on a (100) surface when compared to a distance-based dividing surface.

Balanced mitochondrial and cytosolic translatomes underlie the biogenesis of human respiratory complexes.

BACKGROUND: Oxidative phosphorylation (OXPHOS) complexes consist of nuclear and mitochondrial DNA-encoded subunits. Their biogenesis requires cross-compartment gene regulation to mitigate the accumulation of disproportionate subunits. To determine how human cells coordinate mitochondrial and nuclear gene expression processes, we tailored ribosome profiling for the unique features of the human mitoribosome. RESULTS: We resolve features of mitochondrial translation initiation and identify a small ORF in the 3' UTR of MT-ND5. Analysis of ribosome footprints in five cell types reveals that average mitochondrial synthesis levels correspond precisely to cytosolic levels across OXPHOS complexes, and these average rates reflect the relative abundances of the complexes. Balanced mitochondrial and cytosolic synthesis does not rely on rapid feedback between the two translation systems, and imbalance caused by mitochondrial translation deficiency is associated with the induction of proteotoxicity pathways. CONCLUSIONS: Based on our findings, we propose that human OXPHOS complexes are synthesized proportionally to each other, with mitonuclear balance relying on the regulation of OXPHOS subunit translation across cellular compartments, which may represent a proteostasis vulnerability.

StructRank: a new approach for ligand-based virtual screening.

Screening large libraries of chemical compounds against a biological target, typically a receptor or an enzyme, is a crucial step in the process of drug discovery. Virtual screening (VS) can be seen as a ranking problem which prefers as many actives as possible at the top of the ranking. As a standard, current Quantitative Structure-Activity Relationship (QSAR) models apply regression methods to predict the level of activity for each molecule and then sort them to establish the ranking. In this paper, we propose a top-k ranking algorithm (StructRank) based on Support Vector Machines to solve the early recognition problem directly. Empirically, we show that our ranking approach outperforms not only regression methods but another ranking approach recently proposed for QSAR ranking, RankSVM, in terms of actives found.

Truxillic acid derivatives act as peroxisome proliferator-activated receptor gamma activators.

In previous studies, we identified a truxillic acid derivative as selective activator of the peroxisome proliferator-activated receptor gamma, which is a member of the nuclear receptor family and acts as ligand-activated transcription factor of genes involved in glucose metabolism. Herein we present the structure-activity relationships of 16 truxillic acid derivatives, investigated by a cell-based reporter gene assay guided by molecular docking analysis.

Applicability domains for classification problems: Benchmarking of distance to models for Ames mutagenicity set.

The estimation of accuracy and applicability of QSAR and QSPR models for biological and physicochemical properties represents a critical problem. The developed parameter of "distance to model" (DM) is defined as a metric of similarity between the training and test set compounds that have been subjected to QSAR/QSPR modeling. In our previous work, we demonstrated the utility and optimal performance of DM metrics that have been based on the standard deviation within an ensemble of QSAR models. The current study applies such analysis to 30 QSAR models for the Ames mutagenicity data set that were previously reported within the 2009 QSAR challenge. We demonstrate that the DMs based on an ensemble (consensus) model provide systematically better performance than other DMs. The presented approach identifies 30-60% of compounds having an accuracy of prediction similar to the interlaboratory accuracy of the Ames test, which is estimated to be 90%. Thus, the in silico predictions can be used to halve the cost of experimental measurements by providing a similar prediction accuracy. The developed model has been made publicly available at http://ochem.eu/models/1 .

Benchmark data set for in silico prediction of Ames mutagenicity.

Up to now, publicly available data sets to build and evaluate Ames mutagenicity prediction tools have been very limited in terms of size and chemical space covered. In this report we describe a new unique public Ames mutagenicity data set comprising about 6500 nonconfidential compounds (available as SMILES strings and SDF) together with their biological activity. Three commercial tools (DEREK, MultiCASE, and an off-the-shelf Bayesian machine learner in Pipeline Pilot) are compared with four noncommercial machine learning implementations (Support Vector Machines, Random Forests, k-Nearest Neighbors, and Gaussian Processes) on the new benchmark data set.

Transport of Proteins into Mitochondria.

Mitochondria are essential organelles of eukaryotic cells. They consist of hundreds of different proteins that exhibit crucial activities in respiration, catabolic metabolism and the synthesis of amino acids, lipids, heme and iron-sulfur clusters. With the exception of a handful of hydrophobic mitochondrially encoded membrane proteins, all these proteins are synthesized on cytosolic ribosomes, targeted to receptors on the mitochondrial surface, and transported across or inserted into the outer and inner mitochondrial membrane before they are folded and assembled into their final native structure. This review article provides a comprehensive overview of the mechanisms and components of the mitochondrial protein import systems with a particular focus on recent developments in the field.

An ER surface retrieval pathway safeguards the import of mitochondrial membrane proteins in yeast.

The majority of organellar proteins are translated on cytosolic ribosomes and must be sorted correctly to function. Targeting routes have been identified for organelles such as peroxisomes and the endoplasmic reticulum (ER). However, little is known about the initial steps of targeting of mitochondrial proteins. In this study, we used a genome-wide screen in yeast and identified factors critical for the intracellular sorting of the mitochondrial inner membrane protein Oxa1. The screen uncovered an unexpected path, termed ER-SURF, for targeting of mitochondrial membrane proteins. This pathway retrieves mitochondrial proteins from the ER surface and reroutes them to mitochondria with the aid of the ER-localized chaperone Djp1. Hence, cells use the expanse of the ER surfaces as a fail-safe to maximize productive mitochondrial protein targeting.

Escorted by chaperones: Sti1 helps to usher precursor proteins from the ribosome to mitochondria.

Little is known about factors that interact with mitochondrial precursor proteins in the cytosol. Employing site-specific crosslinking this study identifies chaperones of the Hsp70 and Hsp90 families as well as Sti1 as escorts of cytosolic preproteins. Sti1 presumably helps to hand-over preproteins from Hsp70 to the Hsp90 system and thereby facilitates their binding to TOM receptors on the mitochondrial surface.

Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space.

Simultaneously accurate and efficient prediction of molecular properties throughout chemical compound space is a critical ingredient toward rational compound design in chemical and pharmaceutical industries. Aiming toward this goal, we develop and apply a systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules. These methods range from a simple sum over atoms, to addition of bond energies, to pairwise interatomic force fields, reaching to the more sophisticated machine learning approaches that are capable of describing collective interactions between many atoms or bonds. In the case of equilibrium molecular geometries, even simple pairwise force fields demonstrate prediction accuracy comparable to benchmark energies calculated using density functional theory with hybrid exchange-correlation functionals; however, accounting for the collective many-body interactions proves to be essential for approaching the "holy grail" of chemical accuracy of 1 kcal/mol for both equilibrium and out-of-equilibrium geometries. This remarkable accuracy is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space. In addition, the same representation allows us to predict accurate electronic properties of molecules, such as their polarizability and molecular frontier orbital energies.

Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization Energies.

The accurate and reliable prediction of properties of molecules typically requires computationally intensive quantum-chemical calculations. Recently, machine learning techniques applied to ab initio calculations have been proposed as an efficient approach for describing the energies of molecules in their given ground-state structure throughout chemical compound space (Rupp et al. Phys. Rev. Lett. 2012, 108, 058301). In this paper we outline a number of established machine learning techniques and investigate the influence of the molecular representation on the methods performance. The best methods achieve prediction errors of 3 kcal/mol for the atomization energies of a wide variety of molecules. Rationales for this performance improvement are given together with pitfalls and challenges when applying machine learning approaches to the prediction of quantum-mechanical observables.