Search NASA⌕ Search

SEARCH · Search NASA

Results for “Protein sequencing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

Large language models generate functional protein sequences across diverse families

Deep-learning language models have shown promise in various biotechnological applications, including protein design and engineering. Here, in this paper, we describe ProGen, a language model that can generate protein sequences with a predictable function across large protein families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. The model was trained on 280 million protein sequences from >19,000 families and is augmented with control tags specifying protein properties. ProGen can be further fine-tuned to curated sequences and tags to improve controllable generation performance of proteins from families with sufficient homologous samples. Artificial proteins fine-tuned to five distinct lysozyme families showed similar catalytic efficiencies as natural lysozymes, with sequence identity to natural proteins as low as 31.4%. ProGen is readily adapted to diverse protein families, as we demonstrate with chorismate mutase and malate dehydrogenase.

59 BASIC BIOLOGICAL SCIENCES↗

CAMELOT: A machine learning approach for coarse-grained simulations of aggregation of block-copolymeric protein sequences

We report the development and deployment of a coarse-graining method that is well suited for computer simulations of aggregation and phase separation of protein sequences with block-copolymeric architectures. Our algorithm, named CAMELOT for Coarse-grained simulations Aided by MachinE Learning Optimization and Training, leverages information from converged all atom simulations that is used to determine a suitable resolution and parameterize the coarse-grained model. To parameterize a system-specific coarse-grained model, we use a combination of Boltzmann inversion, non-linear regression, and a Gaussian process Bayesian optimization approach. The accuracy of the coarse-grained model is demonstrated through direct comparisons to results from all atom simulations. We demonstrate the utility of our coarse-graining approach using the block-copolymeric sequence from the exon 1 encoded sequence of the huntingtin protein. This sequence comprises of 17 residues from the N-terminal end of huntingtin (N17) followed by a polyglutamine (polyQ) tract. Simulations based on the CAMELOT approach are used to show that the adsorption and unfolding of the wild type N17 and its sequence variants on the surface of polyQ tracts engender a patchy colloid like architecture that promotes the formation of linear aggregates. These results provide a plausible explanation for experimental observations, which show that N17 accelerates the formation of linear aggregates in block-copolymeric N17-polyQ sequences. The CAMELOT approach is versatile and is generalizable for simulating the aggregation and phase behavior of a range of block-copolymeric protein sequences.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Neural networks to learn protein sequence–function relationships from deep mutational scanning data

Understanding the relationship between protein sequence and function is necessary to design new and useful proteins with applications in bioenergy, medicine, and agriculture. The mapping from sequence to function is tremendously complex because it involves thousands of molecular interactions that are coupled over multiple lengths and timescales. We show that neural networks can learn the sequence–function mapping from large protein datasets. Neural networks are appealing for this task because they can learn complicated relationships from data, make few assumptions about the nature of the sequence–function relationship, and can learn general rules that apply across the length of the protein sequence. We demonstrate that learned models can be applied to design new proteins with properties that exceed natural sequences.

59 BASIC BIOLOGICAL SCIENCES↗

Robust deep learning–based protein sequence design using ProteinMPNN

Although deep learning has revolutionized protein structure prediction, almost all experimentally characterized de novo protein designs have been generated using physically based approaches such as Rosetta. Here, we describe a deep learning–based protein sequence design method, ProteinMPNN, that has outstanding performance in both in silico and experimental tests. On native protein backbones, ProteinMPNN has a sequence recovery of 52.4% compared with 32.9% for Rosetta. The amino acid sequence at different positions can be coupled between single or multiple chains, enabling application to a wide range of current protein design challenges. We demonstrate the broad utility and high accuracy of ProteinMPNN using x-ray crystallography, cryo–electron microscopy, and functional studies by rescuing previously failed designs, which were made using Rosetta or AlphaFold, of protein monomers, cyclic homo-oligomers, tetrahedral nanoparticles, and target-binding proteins.

59 BASIC BIOLOGICAL SCIENCES↗

Clustering huge protein sequence sets in linear time

Metagenomic datasets contain billions of protein sequences that could greatly enhance largescale functional annotation and structure prediction. Utilizing this enormous resource would require reducing its redundancy by similarity clustering. However, clustering hundreds of millions of sequences is impractical using current algorithms because their runtimes scale as the input set size N times the number of clusters K, which is typically of similar order as N, resulting in runtimes that increase almost quadratically with N. We developed Linclust, the first clustering algorithm whose runtime scales as N, independent of K. It can also cluster datasets several times larger than the available main memory. We cluster 1.6 billion metagenomic sequence fragments in 10 h on a single server to 50% sequence identity, >1000 times faster than has been possible before. Linclust will help to unlock the great wealth contained in metagenomic and genomic sequence databases.

59 BASIC BIOLOGICAL SCIENCES↗

Establishing homologies in protein sequences

Computer-based statistical techniques used to determine homologies between proteins occurring in different species are reviewed. The technique is based on comparison of two protein sequences, either by relating all segments of a given length in one sequence to all segments of the second or by finding the best alignment of the two sequences. Approaches discussed include selection using printed tabulations, identification of very similar sequences, and computer searches of a database. The use of the SEARCH, RELATE, and ALIGN programs (Dayhoff, 1979) is explained; sample data are presented in graphs, diagrams, and tables and the construction of scoring matrices is considered.

Dayhoff, M. O.↗

FASTERp: A Feature Array Search Tool for Estimating Resemblance of Protein Sequences

Metagenome sequencing efforts have provided a large pool of billions of genes for identifying enzymes with desirable biochemical traits. However, homology search with billions of genes in a rapidly growing database has become increasingly computationally impractical. Here we present our pilot efforts to develop a novel alignment-free algorithm for homology search. Specifically, we represent individual proteins as feature vectors that denote the presence or absence of short kmers in the protein sequence. Similarity between feature vectors is then computed using the Tanimoto score, a distance metric that can be rapidly computed on bit string representations of feature vectors. Preliminary results indicate good correlation with optimal alignment algorithms (Spearman r of 0.87, ~;;1,000,000 proteins from Pfam), as well as with heuristic algorithms such as BLAST (Spearman r of 0.86, ~;;1,000,000 proteins). Furthermore, a prototype of FASTERp implemented in Python runs approximately four times faster than BLAST on a small scale dataset (~;;1000 proteins). We are optimizing and scaling to improve FASTERp to enable rapid homology searches against billion-protein databases, thereby enabling more comprehensive gene annotation efforts.

59 BASIC BIOLOGICAL SCIENCES↗

Thermal adaptation analyzed by comparison of protein sequences from mesophilic and extremely thermophilic Methanococcus species

The genome sequence of the extremely thermophilic archaeon Methanococcus jannaschii provides a wealth of data on proteins from a thermophile. In this paper, sequences of 115 proteins from M. jannaschii are compared with their homologs from mesophilic Methanococcus species. Although the growth temperatures of the mesophiles are about 50 degrees C below that of M. jannaschii, their genomic G+C contents are nearly identical. The properties most correlated with the proteins of the thermophile include higher residue volume, higher residue hydrophobicity, more charged amino acids (especially Glu, Arg, and Lys), and fewer uncharged polar residues (Ser, Thr, Asn, and Gln). These are recurring themes, with all trends applying to 83-92% of the proteins for which complete sequences were available. Nearly all of the amino acid replacements most significantly correlated with the temperature change are the same relatively conservative changes observed in all proteins, but in the case of the mesophile/thermophile comparison there is a directional bias. We identify 26 specific pairs of amino acids with a statistically significant (P < 0.01) preferred direction of replacement.

NASA Discipline Exobiology↗

Protein sequences and redox titrations indicate that the electron acceptors in reaction centers from heliobacteria are similar to Photosystem I

Photosynthetic reaction centers isolated from Heliobacillus mobilis exhibit a single major protein on SDS-PAGE of 47 000 Mr. Attempts to sequence the reaction center polypeptide indicated that the N-terminus is blocked. After enzymatic and chemical cleavage, four peptide fragments were sequenced from the Heliobacillus mobilis apoprotein. Only one of these sequences showed significant specific similarity to any of the protein and deduced protein sequences in the GenBank data base. This fragment is identical with 56% of the residues, including both cysteines, found in highly conserved region that is proposed to bind iron-sulfur center Fx in the Photosystem I reaction center peptide that is the psaB gene product. The similarity to the psaA gene product in this region is 48%. Redox titrations of laser-flash-induced photobleaching with millisecond decay kinetics on isolated reaction centers from Heliobacterium gestii indicate a midpoint potential of -414 mV with n = 2 titration behavior. In membranes, the behavior is intermediate between n = 1 and n = 2, and the apparent midpoint potential is -444 mV. This is compared to the behavior in Photosystem I, where the intermediate electron acceptor A1, thought to be a phylloquinone molecule, has been proposed to undergo a double reduction at low redox potentials in the presence of viologen redox mediators. These results strongly suggest that the acceptor side electron transfer system in reaction centers from heliobacteria is indeed analogous to that found in Photosystem I. The sequence similarities indicate that the divergence of the heliobacteria from the Photosystem I line occurred before the gene duplication and subsequent divergence that lead to the heterodimeric protein core of the Photosystem I reaction center.

NASA Discipline Number 52-30↗

Front-End Electron Transfer Dissociation Coupled to a 21 Tesla FT-ICR Mass Spectrometer for Intact Protein Sequence Analysis

High resolution mass spectrometry is a key technology for in-depth protein characterization. High-field Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) enables high-level interrogation of intact proteins in the most detail to date. However, an appropriate complement of fragmentation technologies must be paired with FTMS to provide comprehensive sequence coverage, as well as characterization of sequence variants, and post-translational modifications. Here we describe the integration of front-end electron transfer dissociation (FETD) with a custom-built 21 tesla FT-ICR mass spectrometer, which yields unprecedented sequence coverage for proteins ranging from 2.8 to 29 kDa, without the need for extensive spectral averaging (e.g., ~60% sequence coverage for apo-myoglobin with four averaged acquisitions). The system is equipped with a multipole storage device separate from the ETD reaction device, which allows accumulation of multiple ETD fragment ion fills. Consequently, an optimally large product ion population is accumulated prior to transfer to the ICR cell for mass analysis, which improves mass spectral signal-to-noise ratio, dynamic range, and scan rate. We find a linear relationship between protein molecular weight and minimum number of ETD reaction fills to achieve optimum sequence coverage, thereby enabling more efficient use of instrument data acquisition time. Finally, real-time scaling of the number of ETD reactions fills during method-based acquisition is shown, and the implications for LC-MS/MS top-down analysis are discussed. .

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Plant sulfate transporter protein sequences for phylogenetic analysis

Sulfur is an essential macronutrient that supports plant growth, development, and responses to environmental stress. Sulfate is the predominant inorganic form of sulfur in soils, and its uptake by roots and translocation to shoots are facilitated by the sulfate transporter (SULTR) family of proteins. Although the first plant SULTR gene was identified nearly three decades ago, several subfamily members, particularly those in the expansive and angiosperm-specific SULTR3 group, remain poorly characterized. To support comprehensive phylogenetic and sequence-based analyses, we compiled a curated dataset of 262 SULTR protein sequences from 22 plant species spanning the evolutionary breadth of land plants. This collection includes representatives from two basal lineages, two early-divergent angiosperms, six monocots, and ten dicots. All sequences were extracted from genome assemblies available in Phytozome v13 (Joint Genome Institute) and manually curated, with cross-referencing to additional databases such as NCBI when needed. This dataset provides a valuable resource for reconstructing the evolutionary history of the SULTR family, with particular emphasis on the diversification of SULTR3 transporters in flowering plants. This resource may also support functional annotation, comparative genomics, and structural modeling of sulfate transport proteins.

CBI↗

Snekmer: a scalable pipeline for protein sequence fingerprinting based on amino acid recoding

Abstract Motivation The vast expansion of sequence data generated from single organisms and microbiomes has precipitated the need for faster and more sensitive methods to assess evolutionary and functional relationships between proteins. Representing proteins as sets of short peptide sequences (kmers) has been used for rapid, accurate classification of proteins into functional categories; however, this approach employs an exact-match methodology and thus may be limited in terms of sensitivity and coverage. We have previously used similarity groupings, based on the chemical properties of amino acids, to form reduced character sets and recode proteins. This amino acid recoding (AAR) approach simplifies the construction of protein representations in the form of kmer vectors, which can link sequences with distant sequence similarity and provide accurate classification of problematic protein families. Results Here, we describe Snekmer, a software tool for recoding proteins into AAR kmer vectors and performing either (i) construction of supervised classification models trained on input protein families or (ii) clustering for de novo determination of protein families. We provide examples of the operation of the tool against a set of nitrogen cycling families originally collected using both standard hidden Markov models and a larger set of proteins from Uniprot and demonstrate that our method accurately differentiates these sequences in both operation modes. Availability and implementation Snekmer is written in Python using Snakemake. Code and data used in this article, along with tutorial notebooks, are available at http://github.com/PNNL-CompBio/Snekmer under an open-source BSD-3 license. Supplementary information Supplementary data are available at Bioinformatics Advances online.

59 BASIC BIOLOGICAL SCIENCES↗

Characterization of bud emergence 46 (BEM46) protein: Sequence, structural, phylogenetic and subcellular localization analyses

Highlights: •All eukaryotes have at least a single copy of a bem46 ortholog. •The catalytic triad of BEM46 is illustrated using sequence and structural analysis. •We identified indels in the conserved domain of BEM46 protein. •Localization studies of BEM46 protein were carried out using GFP-fusion tagging. -- Abstract: The bud emergence 46 (BEM46) protein from Neurospora crassa belongs to the α/β-hydrolase superfamily. Recently, we have reported that the BEM46 protein is localized in the perinuclear ER and also forms spots close by the plasma membrane. The protein appears to be required for cell type-specific polarity formation in N. crassa. Furthermore, initial studies suggested that the BEM46 amino acid sequence is conserved in eukaryotes and is considered to be one of the widespread conserved “known unknown” eukaryotic genes. This warrants for a comprehensive phylogenetic analysis of this superfamily to unravel origin and molecular evolution of these genes in different eukaryotes. Herein, we observe that all eukaryotes have at least a single copy of a bem46 ortholog. Upon scanning of these proteins in various genomes, we find that there are expansions leading into several paralogs in vertebrates. Usingcomparative genomic analyses, we identified insertion/deletions (indels) in the conserved domain of BEM46 protein, which allow to differentiate fungal classes such as ascomycetes from basidiomycetes. We also find that exonic indels are able to differentiate BEM46 homologs of different eukaryotic lineage. Furthermore, we unravel that BEM46 protein from N. crassa possess a novel endoplasmic-retention signal (PEKK) using GFP-fusion tagging experiments. We propose that three residues namely a serine 188S, a histidine 292H and an aspartic acid 262D are most critical residues, forming a catalytic triad in BEM46 protein from N. crassa. We carried out a comprehensive study on bem46 genes from a molecular evolution perspective with combination of functional analyses. The evolutionary history of BEM46 proteins is characterized by exonic indels in lineage specific manner.

60 APPLIED LIFE SCIENCES↗

Reinspection of a Clinical Proteomics Tumor Analysis Consortium (CPTAC) Dataset with Cloud Computing Reveals Abundant Post-Translational Modifications and Protein Sequence Variants

The Clinical Proteomic Tumor Analysis Consortium (CPTAC) has provided some of the most in-depth analyses of the phenotypes of human tumors ever constructed. Today, the majority of proteomic data analysis is still performed using software housed on desktop computers which limits the number of sequence variants and post-translational modifications that can be considered. The original CPTAC studies limited the search for PTMs to only samples that were chemically enriched for those modified peptides. Similarly, the only sequence variants considered were those with strong evidence at the exon or transcript level. In this multi-institutional collaborative reanalysis, we utilized unbiased protein databases containing millions of human sequence variants in conjunction with hundreds of common post-translational modifications. Using these tools, we identified tens of thousands of high-confidence PTMs and sequence variants. We identified 4132 phosphorylated peptides in nonenriched samples, 93% of which were confirmed in the samples which were chemically enriched for phosphopeptides. In addition, our results also cover 90% of the high-confidence variants reported by the original proteogenomics study, without the need for sample specific next-generation sequencing. Finally, we report fivefold more somatic and germline variants that have an independent evidence at the peptide level, including mutations in ERRB2 and BCAS1. In this reanalysis of CPTAC proteomic data with cloud computing, we present an openly available and searchable web resource of the highest-coverage proteomic profiling of human tumors described to date.

60 APPLIED LIFE SCIENCES↗

Material lessons of biology: structure function studies of protein sequences involved in inorganic composite material formation. Final Technical Report

This focus of this grant is to study the interrelationships between proteins that form the high performance bioceramic nacre layer of the mollusc shell. In particular during this grant period we investigated (1) the interactivity between two framework nacre proteins, n16.3 and Pif97 of the Japanese pearl oyster, Pinctada fucata, and (2) the incorporation and subsequent material property modifications that accompany the intracrystalline incorporation of two nacre proteins, AP7 (H. rufescens) and n16.3 (P. fucata).

36 MATERIALS SCIENCE↗

EpitopeVec: linear epitope prediction using deep protein sequence embeddings

Motivation: B-cell epitopes (BCEs) play a pivotal role in the development of peptide vaccines, immuno-diagnostic reagents and antibody production, and thus in infectious disease prevention and diagnostics in general. Experimental methods used to determine BCEs are costly and time-consuming. Therefore, it is essential to develop computational methods for the rapid identification of BCEs. Although several computational methods have been developed for this task, generalizability is still a major concern, where cross-testing of the classifiers trained and tested on different datasets has revealed accuracies of 51–53%. Results: We describe a new method called EpitopeVec, which uses a combination of residue properties, modified antigenicity scales, and protein language model-based representations (protein vectors) as features of peptides for linear BCE predictions. Extensive benchmarking of EpitopeVec and other state-of-the-art methods for linear BCE prediction on several large and small datasets, as well as cross-testing, demonstrated an improvement in the performance of EpitopeVec over other methods in terms of accuracy and area under the curve. As the predictive performance depended on the species origin of the respective antigens (viral, bacterial and eukaryotic), we also trained our method on a large viral dataset to create a dedicated linear viral BCE predictor with improved cross-testing performance.

59 BASIC BIOLOGICAL SCIENCES↗