Better together: Elements of successful scientific software development in a distributed collaborative community.

Many scientific disciplines rely on computational methods for data analysis, model generation, and prediction. Implementing these methods is often accomplished by researchers with domain expertise but without formal training in software engineering or computer science. This arrangement has led to underappreciation of sustainability and maintainability of scientific software tools developed in academic environments. Some software tools have avoided this fate, including the scientific library Rosetta. We use this software and its community as a case study to show how modern software development can be accomplished successfully, irrespective of subject area. Rosetta is one of the largest software suites for macromolecular modeling, with 3.1 million lines of code and many state-of-the-art applications. Since the mid 1990s, the software has been developed collaboratively by the RosettaCommons, a community of academics from over 60 institutions worldwide with diverse backgrounds including chemistry, biology, physiology, physics, engineering, mathematics, and computer science. Developing this software suite has provided us with more than two decades of experience in how to effectively develop advanced scientific software in a global community with hundreds of contributors. Here we illustrate the functioning of this development community by addressing technical aspects (like version control, testing, and maintenance), community-building strategies, diversity efforts, software dissemination, and user support. We demonstrate how modern computational research can thrive in a distributed collaborative community. The practices described here are independent of subject area and can be readily adopted by other software development communities.

A microbial sensor for organophosphate hydrolysis exploiting an engineered specificity switch in a transcription factor.

A whole-cell biosensor utilizing a transcription factor (TF) is an effective tool for sensitive and selective detection of specialty chemicals or anthropogenic molecules, but requires access to an expanded repertoire of TFs. Using homology modeling and ligand docking for binding pocket identification, assisted by conservative mutations in the pocket, we engineered a novel specificity in an Acinetobacter TF, PobR, to 'sense' a chemical p-nitrophenol (pNP) and measured the response via a fluorescent protein reporter expressed from a PobR promoter. Out of 10(7) variants of PobR, four were active when dosed with pNP, with two mutants showing a specificity switch from the native effector 4-hydroxybenzoate (4HB). One of the mutants, pNPmut1 was then used to create a smart microbial cell responding to pNP production from hydrolysis of an insecticide, paraoxon, in a coupled assay involving phosphotriesterase (PTE) enzyme expressed from a separate promoter. We show the fluorescence of the cells correlated with the catalytic efficiency of the PTE variant expressed in each cell. High selectivity between similar molecules (4HB versus pNP), high sensitivity for pNP detection (∼2 µM) and agreement of apo- and holo-structures of PobR scaffold with predetermined computational models are other significant results presented in this work.

Rosetta comparative modeling for library design: Engineering alternative inducer specificity in a transcription factor.

Structure-based rational mutagenesis for engineering protein functionality has been limited by the scarcity and difficulty of obtaining crystal structures of desired proteins. On the other hand, when high-throughput selection is possible, directed evolution-based approaches for gaining protein functionalities have been random and fortuitous with limited rationalization. We combine comparative modeling of dimer structures, ab initio loop reconstruction, and ligand docking to select positions for mutagenesis to create a library focused on the ligand-contacting residues. The rationally reduced library requirement enabled conservative control of the substitutions by oligonucleotide synthesis and bounding its size within practical transformation efficiencies (∼ 10(7) variants). This rational approach was successfully applied on an inducer-binding domain of an Acinetobacter transcription factor (TF), pobR, which shows high specificity for natural effector molecule, 4-hydroxy benzoate (4HB), but no native response to 3,4-dihydroxy benzoate (34DHB). Selection for mutants with high transcriptional induction by 34DHB was carried out at the single-cell level under flow cytometry (via green fluorescent protein expression under the control of pobR promoter). Critically, this selection protocol allows both selection for induction and rejection of constitutively active mutants. In addition to gain-of-function for 34DHB induction, the selected mutants also showed enhanced sensitivity and response for 4HB (native inducer) while no sensitivity was observed for a non-targeted but chemically similar molecule, 2-hydroxy benzoate (2HB). This is unique application of the Rosetta modeling protocols for library design to engineer a TF. Our approach extends applicability of the Rosetta redesign protocol into regimes without a priori precision structural information.

Protein remote homology detection and structural alignment using deep learning.

Exploiting sequence-structure-function relationships in biotechnology requires improved methods for aligning proteins that have low sequence similarity to previously annotated proteins. We develop two deep learning methods to address this gap, TM-Vec and DeepBLAST. TM-Vec allows searching for structure-structure similarities in large sequence databases. It is trained to accurately predict TM-scores as a metric of structural similarity directly from sequence pairs without the need for intermediate computation or solution of structures. Once structurally similar proteins have been identified, DeepBLAST can structurally align proteins using only sequence information by identifying structurally homologous regions between proteins. It outperforms traditional sequence alignment methods and performs similarly to structure-based alignment methods. We show the merits of TM-Vec and DeepBLAST on a variety of datasets, including better identification of remotely homologous proteins compared with state-of-the-art sequence alignment and structure prediction methods.

Label-free affinity screening, design and synthesis of inhibitors targeting the Mycobacterium tuberculosis L-alanine dehydrogenase.

The ability of Mycobacterium tuberculosis (Mtb) to persist in its host may enable an evolutionary advantage for drug resistant variants to emerge. A potential strategy to prevent persistence and gain drug efficacy is to directly target the activity of enzymes that are crucial for persistence. We present a method for expedited discovery and structure-based design of lead compounds by targeting the hypoxia-associated enzyme L-alanine dehydrogenase (AlaDH). Biochemical and structural analyses of AlaDH confirmed binding of nucleoside derivatives and showed a site adjacent to the nucleoside binding pocket that can confer specificity to putative inhibitors. Using a combination of dye-ligand affinity chromatography, enzyme kinetics and protein crystallographic studies, we show the development and validation of drug prototypes. Crystal structures of AlaDH-inhibitor complexes with variations at the N6 position of the adenyl-moiety of the inhibitor provide insight into the molecular basis for the specificity of these compounds. We describe a drug-designing pipeline that aims to block Mtb to proliferate upon re-oxygenation by specifically blocking NAD accessibility to AlaDH. The collective approach to drug discovery was further evaluated through in silico analyses providing additional insight into an efficient drug development strategy that can be further assessed with the incorporation of in vivo studies.

Emergence of symmetry in homooligomeric biological assemblies.

Naturally occurring homooligomeric protein complexes exhibit striking internal symmetry. The evolutionary origins of this symmetry have been the subject of considerable speculation; proposals for the advantages associated with symmetry include greater folding efficiency, reduced aggregation, amenability to allosteric regulation, and greater adaptability. An alternative possibility stems from the idea that to contribute to fitness, and hence be subject to evolutionary optimization, a complex must be significantly populated, which implies that the interaction energy between monomers in the ancestors of modern-day complexes must have been sufficient to at least partially overcome the entropic cost of association. Here, we investigate the effects of this bias toward very-low-energy complexes on the distribution of symmetry in primordial homooligomers modeled as randomly interacting pairs of monomers. We demonstrate quantitatively that a bias toward very-low-energy complexes can result in the emergence of symmetry from random ensembles in which the overall frequency of symmetric complexes is vanishingly small. This result is corroborated by using explicit protein-protein docking calculations to generate ensembles of randomly docked complexes: the fraction of these that are symmetric increases from 0.02% in the overall population to >50% in very low energy subpopulations.

Engineered pH-Sensitive Protein G/IgG Interaction.

While natural protein-protein interactions have evolved to be induced by complex stimuli, rational design of interactions that can be switched-on-demand still remain challenging in the protein design world. Here, we demonstrate that a computationally redesigned natural interface for improved binding affinity could further be mutated to adopt a pH switchable interaction. The redesigned interface of Protein G/human IgG Fc domain (referred to as PrG/hIgG), when incorporated with histidine and glutamic acid on PrG (PrG-EHHE), showed a switch in binding affinity by 50-fold when the pH was altered from mild acidic to mild basic. The wild-type (WT) interface showed a negligible switch. The overall binding affinity under mild acidic pH for PrG-EHHE outperformed the wild-type PrG (PrG-WT) interaction. The new reagent PrG-EHHE can be revolutionary in IgG purification, since the standard method of using an extreme acidic pH for elution can be circumvented.

Conformational control via sequence for a heteropeptoid in water: coupled NMR and Rosetta modelling.

We report a critical advance in the generation and characterization of peptoid hetero-oligomers. A library of sub-monomers with amine and carboxylate side-chains are combined in different sequences using microwave-assisted synthesis. Their sequence-structure propensity is confirmed by circular dichroism, and conformer subtypes are enumerated by NMR. Biasing the ψ-angle backbone to trans (180°) in Monte Carlo modelling favors i to i + 3 naphthyl-naphthyl stacking, and matches experimental ensemble distributions. Taken together, high-yield synthesis of heterooligomers and NMR with structure prediction enables rapid determination of sequences that induce secondary structural propensities for predictive design of hydrophilic peptidomimetic foldamers and their future libraries.

Superfamily assignments for the yeast proteome through integration of structure prediction with the gene ontology.

Saccharomyces cerevisiae is one of the best-studied model organisms, yet the three-dimensional structure and molecular function of many yeast proteins remain unknown. Yeast proteins were parsed into 14,934 domains, and those lacking sequence similarity to proteins of known structure were folded using the Rosetta de novo structure prediction method on the World Community Grid. This structural data was integrated with process, component, and function annotations from the Saccharomyces Genome Database to assign yeast protein domains to SCOP superfamilies using a simple Bayesian approach. We have predicted the structure of 3,338 putative domains and assigned SCOP superfamily annotations to 581 of them. We have also assigned structural annotations to 7,094 predicted domains based on fold recognition and homology modeling methods. The domain predictions and structural information are available in an online database at http://rd.plos.org/10.1371_journal.pbio.0050076_01.

Smart Microbial Cells Couple Catalysis and Sensing to Provide High-Throughput Selection of an Organophosphate Hydrolase.

Enzyme engineering for gain of function requires navigating a large combinatorial sequence space efficiently. Typically, many mutations are needed to get significant improvements, while a single "bad" mutation can inactivate the enzyme. To establish high-throughput screening and achieve enhanced resolution between two variants, genetic libraries of the organophosphate hydrolase enzyme paraoxonase 1 (PON1) were rapidly screened via an engineered positive-feedback circuit: a p-nitrophenol (PNP)-specific transcription factor (TF) regulated expression of PON1, which catalyzed paraoxon breakdown and PNP production. Rare active mutant colonies, picked by simple visual fluorescence of a PON1-green fluorescent protein (GFP) fusion, were characterized. In a single screening round, high (library-scale) throughput enabled the discovery of enhanced paraoxon degradation activity in PON1, including structurally unexpected mutations.

Sensor-Enabled Alleviation of Product Inhibition in Chorismate Pyruvate-Lyase.

Product inhibition is a frequent bottleneck in industrial enzymes, and testing mutations to alleviate product inhibition via traditional methods remains challenging as many variants need to be tested against multiple substrate and product concentrations. Further, traditional screening methods are conducted in vitro, and resulting enzyme variants may perform differently in vivo in the context of whole-cell metabolism and regulation. In this study, we address these two problems by establishing a high-throughput screening method to alleviate product inhibition in an industrially relevant enzyme, chorismate pyruvate-lyase (UbiC). First, we engineered a highly specific, genetically encoded biosensor for 4-hydroxybenzoate (4HB) in an industrially relevant host, Pseudomonas putida KT2440. We subsequently applied the biosensor to detect the activity of a heterologously expressed UbiC that converts chorismate into 4HB and pyruvate. By using benzoate as a product surrogate that inhibits UbiC without activating the biosensor, we were able to efficiently create and screen a diversified library for UbiC variants with reduced product inhibition. Introduction of the improved UbiC enzyme variant into an experimental production strain for the industrial precursor cis,cis-muconic acid (muconate), enabled a >2-fold yield improvement for glucose to muconate conversion when the new UbiC variant was expressed from a plasmid and a 60% yield increase when the same UbiC variant was genomically integrated into the strain. Overall, this work demonstrates that by coupling a library of enzyme variants to whole-cell catalysis and biosensing, variants with reduced product inhibition can be identified, and that this improved enzyme can result in increased titers of a downstream molecule of interest.

Deriving topology and sequence alignment for the helix skeleton in low-resolution protein density maps.

Cryoelectron microscopy (cryoEM) is an experimental technique to determine the three-dimensional (3D) structure of large protein complexes. Currently, this technique is able to generate protein density maps at 6-9 A resolution, at which the skeleton of the structure (which is composed of alpha-helices and beta-sheets) can be visualized. As a step towards predicting the entire backbone of the protein from the protein density map, we developed a method to predict the topology and sequence alignment for the skeleton helices. Our method combines the geometrical information of the skeleton helices with the Rosetta ab initio structure prediction method to derive a consensus topology and sequence alignment for the skeleton helices. We tested the method with 60 proteins. For 45 proteins, the majority of the skeleton helices were assigned a correct topology from one of our top ten predictions. The offsets of the alignment for most of the assigned helices were within +/-2 amino acids in the sequence. We also analyzed the use of the skeleton helices as a clustering tool for the decoy structures generated by Rosetta. Our comparison suggests that the topology clustering is a better method than a general overlap clustering method to enrich the ranking of decoys, particularly when the decoy pool is small.

A protocatechuate biosensor for Pseudomonas putida KT2440 via promoter and protein evolution.

Robust fluorescence-based biosensors are emerging as critical tools for high-throughput strain improvement in synthetic biology. Many biosensors are developed in model organisms where sophisticated synthetic biology tools are also well established. However, industrial biochemical production often employs microbes with phenotypes that are advantageous for a target process, and biosensors may fail to directly transition outside the host in which they are developed. In particular, losses in sensitivity and dynamic range of sensing often occur, limiting the application of a biosensor across hosts. Here we demonstrate the optimization of an Escherichia coli-based biosensor in a robust microbial strain for the catabolism of aromatic compounds, Pseudomonas putida KT2440, through a generalizable approach of modulating interactions at the protein-DNA interface in the promoter and the protein-protein dimer interface. The high-throughput biosensor optimization approach demonstrated here is readily applicable towards other allosteric regulators.

Rapid Thermostabilization of Bacillus thuringiensis Serovar Konkukian 97-27 Dehydroshikimate Dehydratase through a Structure-Based Enzyme Design and Whole Cell Activity Assay.

Thermostabilization of an enzyme with complete retention of catalytic efficiency was demonstrated on recombinant 3-dehydroshikimate dehydratase (DHSase or wtAsbF) from Bacillus thuringiensis serovar konkukian 97-27 (hereafter, B. thuringiensis 97-27). The wtAsbF is relatively unstable at 37 °C, in vitro (t1/237 = 15 min), in the absence of divalent metal. We adopted a structure-based design to identify stabilizing mutations and created a combinatorial library based upon predicted mutations at specific locations on the enzyme surface. A diversified asbF library (∼2000 variants) was expressed in E. coli harboring a green fluorescent protein (GFP) reporter system linked to the product of wtAsbF activity (3,4-dihydroxybenzoate, DHB). Mutations detrimental to DHSase function were rapidly eliminated using a high throughput fluorescence activated cell sorting (FACS) approach. After a single sorting round and heat screen at 50 °C, a triple AsbF mutant (Mut1), T61N, H135Y, and H257P, was isolated and characterized. The half-life of Mut1 at 37 °C was >10-fold higher than the wtAsbF (t1/237 = 169 min). Further, the second-order rate constants for both wtAsbF and Mut1 were approximately equal (9.9 × 105 M-1 s-1, 7.8 × 105 M-1 s-1, respectively), thus demonstrating protein thermostability did not come at the expense of enzyme thermophilicity. In addition, in vivo overexpression of Mut1 in E. coli resulted in a ∼60-fold increase in functional enzyme when compared to the wild-type enzyme under the identical expression conditions. Finally, overexpression of the thermostable AsbF resulted in an approximate 80-120% increase in DHB accumulation in the media relative to the wild-type enzyme.

De novo prediction of three-dimensional structures for major protein families.

We use the Rosetta de novo structure prediction method to produce three-dimensional structure models for all Pfam-A sequence families with average length under 150 residues and no link to any protein of known structure. To estimate the reliability of the predictions, the method was calibrated on 131 proteins of known structure. For approximately 60% of the proteins one of the top five models was correctly predicted for 50 or more residues, and for approximately 35%, the correct SCOP superfamily was identified in a structure-based search of the Protein Data Bank using one of the models. This performance is consistent with results from the fourth critical assessment of structure prediction (CASP4). Correct and incorrect predictions could be partially distinguished using a confidence function based on a combination of simulation convergence, protein length and the similarity of a given structure prediction to known protein structures. While the limited accuracy and reliability of the method precludes definitive conclusions, the Pfam models provide the only tertiary structure information available for the 12% of publicly available sequences represented by these large protein families.

MAMMOTH (matching molecular models obtained from theory): an automated method for model comparison.

Advances in structural genomics and protein structure prediction require the design of automatic, fast, objective, and well benchmarked methods capable of comparing and assessing the similarity of low-resolution three-dimensional structures, via experimental or theoretical approaches. Here, a new method for sequence-independent structural alignment is presented that allows comparison of an experimental protein structure with an arbitrary low-resolution protein tertiary model. The heuristic algorithm is given and then used to show that it can describe random structural alignments of proteins with different folds with good accuracy by an extreme value distribution. From this observation, a structural similarity score between two proteins or two different conformations of the same protein is derived from the likelihood of obtaining a given structural alignment by chance. The performance of the derived score is then compared with well established, consensus manual-based scores and data sets. We found that the new approach correlates better than other tools with the gold standard provided by a human evaluator. Timings indicate that the algorithm is fast enough for routine use with large databases of protein models. Overall, our results indicate that the new program (MAMMOTH) will be a good tool for protein structure comparisons in structural genomics applications. MAMMOTH is available from our web site at http://physbio.mssm.edu/~ortizg/.

Modeling structurally variable regions in homologous proteins with rosetta.

A major limitation of current comparative modeling methods is the accuracy with which regions that are structurally divergent from homologues of known structure can be modeled. Because structural differences between homologous proteins are responsible for variations in protein function and specificity, the ability to model these differences has important functional consequences. Although existing methods can provide reasonably accurate models of short loop regions, modeling longer structurally divergent regions is an unsolved problem. Here we describe a method based on the de novo structure prediction algorithm, Rosetta, for predicting conformations of structurally divergent regions in comparative models. Initial conformations for short segments are selected from the protein structure database, whereas longer segments are built up by using three- and nine-residue fragments drawn from the database and combined by using the Rosetta algorithm. A gap closure term in the potential in combination with modified Newton's method for gradient descent minimization is used to ensure continuity of the peptide backbone. Conformations of variable regions are refined in the context of a fixed template structure using Monte Carlo minimization together with rapid repacking of side-chains to iteratively optimize backbone torsion angles and side-chain rotamers. For short loops, mean accuracies of 0.69, 1.45, and 3.62 A are obtained for 4, 8, and 12 residue loops, respectively. In addition, the method can provide reasonable models of conformations of longer protein segments: predicted conformations of 3A root-mean-square deviation or better were obtained for 5 of 10 examples of segments ranging from 13 to 34 residues. In combination with a sequence alignment algorithm, this method generates complete, ungapped models of protein structures, including regions both similar to and divergent from a homologous structure. This combined method was used to make predictions for 28 protein domains in the Critical Assessment of Protein Structure 4 (CASP 4) and 59 domains in CASP 5, where the method ranked highly among comparative modeling and fold recognition methods. Model accuracy in these blind predictions is dominated by alignment quality, but in the context of accurate alignments, long protein segments can be accurately modeled. Notably, the method correctly predicted the local structure of a 39-residue insertion into a TIM barrel in CASP 5 target T0186.

Carbon sequestration in Synechococcus Sp.: from molecular machines to hierarchical modeling.

The U.S. Department of Energy recently announced the first five grants for the Genomes to Life (GTL) Program. The goal of this program is to "achieve the most far-reaching of all biological goals: a fundamental, comprehensive, and systematic understanding of life." While more information about the program can be found at the GTL website (www.doegenomestolife.org), this paper provides an overview of one of the five GTL projects funded, "Carbon Sequestration in Synechococcus Sp.: From Molecular Machines to Hierarchical Modeling." This project is a combined experimental and computational effort emphasizing developing, prototyping, and applying new computational tools and methods to elucidate the biochemical mechanisms of the carbon sequestration of Synechococcus Sp., an abundant marine cyanobacteria known to play an important role in the global carbon cycle. Understanding, predicting, and perhaps manipulating carbon fixation in the oceans has long been a major focus of biological oceanography and has more recently been of interest to a broader audience of scientists and policy makers. It is clear that the oceanic sinks and sources of CO(2) are important terms in the global environmental response to anthropogenic atmospheric inputs of CO(2) and that oceanic microorganisms play a key role in this response. However, the relationship between this global phenomenon and the biochemical mechanisms of carbon fixation in these microorganisms is poorly understood. The project includes five subprojects: an experimental investigation, three computational biology efforts, and a fifth which deals with addressing computational infrastructure challenges of relevance to this project and the Genomes to Life program as a whole. Our experimental effort is designed to provide biology and data to drive the computational efforts and includes significant investment in developing new experimental methods for uncovering protein partners, characterizing protein complexes, identifying new binding domains. We will also develop and apply new data measurement and statistical methods for analyzing microarray experiments. Our computational efforts include coupling molecular simulation methods with knowledge discovery from diverse biological data sets for high-throughput discovery and characterization of protein-protein complexes and developing a set of novel capabilities for inference of regulatory pathways in microbial genomes across multiple sources of information through the integration of computational and experimental technologies. These capabilities will be applied to Synechococcus regulatory pathways to characterize their interaction map and identify component proteins in these pathways. We will also investigate methods for combining experimental and computational results with visualization and natural language tools to accelerate discovery of regulatory pathways. Furthermore, given that the ultimate goal of this effort is to develop a systems-level of understanding of how the Synechococcus genome affects carbon fixation at the global scale, we will develop and apply a set of tools for capturing the carbon fixation behavior of complex of Synechococcus at different levels of resolution. Finally, because the explosion of data being produced by high-throughput experiments requires data analysis and models which are more computationally complex, more heterogeneous, and require coupling to ever increasing amounts of experimentally obtained data in varying formats, we have also established a companion computational infrastructure to support this effort as well as the Genomes to Life program as a whole.

An improved Protein G with higher affinity for human/rabbit IgG Fc domains exploiting a computationally designed polar network.

Protein G is an IgG binding protein that has been widely exploited for biotechnological purposes. Rosetta protein modeling identified a set of favorable polar mutations in Protein G, at its binding interface with the Fc domain of Immunoglobulin G, that were predicted to increase the stability and tighten the binding relative to native Protein G, with only a minor perturbation of the binding mode seen in the crystal structure. This triple mutant was synthesized and evaluated experimentally. Relative to the native protein G, the mutant showed a 3.5-fold enhancement in display level on the surface of yeast and a 5-fold tighter molar affinity for rabbit and human IgG. We attribute the improved affinity to a network of hydrogen bonds exploiting specific polar groups on human and rabbit Fc. The relative specificity increased as well since there was little affinity enhancement for goat and mouse Fc, while the affinity for rat Fc was poorer by half. This designed Protein G will be useful in biotechnological applications as a recombinant protein, where its improved affinity, display and specificity will increase antibody capture sensitivity and capacity. Furthermore, the display of this protein on the surface of yeast introduces the concept of the use of yeast as an affinity matrix.

3D structure analysis of PAKs: A clue to the rational design for affinity reagents and blockers.

The p21-activated kinase (PAK) family plays a versatile role in cell signaling by forming a hub of interactions. PAKs bind the GTPases like RAC and CDC42. Their proline-rich motifs bind SH3 adaptor proteins such as PIX and NCK. PAKs display nuclear localization signal sites and a potential Integrin binding site. No fully complete structure of the PAKs has been published; partial 3D structures of the PAK family kinases include portions of the auto-inhibited PAK1, GTPase bound to small peptides from PAKs, and the kinase domains from PAK1 and PAK4-6 (with small ligands in a few cases). This review focuses on exploring the intermolecular interaction regions in these 3D structures and we offer insights on the missing regions in crystal structure of the auto-inhibited PAK1. Understanding and modulation of PAK intermolecular interactions can pave the way for PAK blockers and biosensors.