Function Prediction Algorithms Research Articles

SAFPred: Synteny-aware gene function prediction for bacteria using protein embeddings.

Today, we know the function of only a small fraction of the protein sequences predicted from genomic data. This problem is even more salient for bacteria, which represent some of the most phylogenetically and metabolically diverse taxa on Earth. This low rate of bacterial gene annotation is compounded by the fact that most function prediction algorithms have focused on eukaryotes, and conventional annotation approaches rely on the presence of similar sequences in existing databases. However, often there are no such sequences for novel bacterial proteins. Thus, we need improved gene function prediction methods tailored for bacteria. Recently, transformer-based language models-adopted from the natural language processing field-have been used to obtain new representations of proteins, to replace amino acid sequences. These representations, referred to as protein embeddings, have shown promise for improving annotation of eukaryotes, but there have been only limited applications on bacterial genomes. To predict gene functions in bacteria, we developed SAFPred, a novel synteny-aware gene function prediction tool based on protein embeddings from state-of-the-art protein language models. SAFpred also leverages the unique operon structure of bacteria through conserved synteny. SAFPred outperformed both conventional sequence-based annotation methods and state-of-the-art methods on multiple bacterial species, including for distant homolog detection, where the sequence similarity to the proteins in the training set was as low as 40%. Using SAFPred to identify gene functions across diverse enterococci, of which some species are major clinical threats, we identified 11 previously unrecognized putative novel toxins, with potential significance to human and animal health. https://github.com/AbeelLab/safpred. Supplementary data are available at Bioinformatics online.

De Novo Multi-Omics Pathway Analysis Designed for Prior Data Independent Inference of Cell Signaling Pathways

New tools for cell signaling pathway inference from multi-omics data that are independent of previous knowledge are needed. Here, we propose a new de novo method, the de novo multi-omics pathway analysis (DMPA), to model and combine omics data into network modules and pathways. DMPA was validated with published omics data and was found accurate in discovering reported molecular associations in transcriptome, interactome, phosphoproteome, methylome, and metabolomics data and signaling pathways in multi-omics data. DMPA was benchmarked against module discovery and multi-omics integration methods and outperformed previous methods in module and pathway discovery especially when applied to datasets with relatively low sample sizes. Transcription factor, kinase, subcellular location, and function prediction algorithms were devised for transcriptome, phosphoproteome, and interactome modules and pathways, respectively. To apply DMPA in a biologically relevant context, interactome, phosphoproteome, transcriptome, and proteome data were collected from analyses carried out using melanoma cells to address gamma-secretase cleavage-dependent signaling characteristics of the receptor tyrosine kinase TYRO3. The pathways modeled with DMPA reflected the predicted function and its direction in validation experiments.

In vivo versus in silico assessment of potentially pathogenic missense variants in human reproductive genes

Infertility is a heterogeneous condition, with genetic causes thought to underlie a substantial fraction of cases. Genome sequencing is becoming increasingly important for genetic diagnosis of diseases including idiopathic infertility; however, most rare or minor alleles identified in patients are variants of uncertain significance (VUS). Interpreting the functional impacts of VUS is challenging but profoundly important for clinical management and genetic counseling. To determine the consequences of these variants in key fertility genes, we functionally evaluated 11 missense variants in the genes ANKRD31, BRDT, DMC1, EXO1, FKBP6, MCM9, M1AP, MEI1, MSH4 and SEPT12 by generating genome-edited mouse models. Nine variants were classified as deleterious by most functional prediction algorithms, and two disrupted a protein-protein interaction (PPI) in the yeast two hybrid (Y2H) assay. Though these genes are essential for normal meiosis or spermiogenesis in mice, only one variant, observed in the MCM9 gene of a male infertility patient, compromised fertility or gametogenesis in the mouse models. To explore the disconnect between predictions and outcomes, we compared pathogenicity calls of missense variants made by ten widely used algorithms to 1) those annotated in ClinVar and 2) those evaluated in mice. All the algorithms performed poorly in terms of predicting the effects of human missense variants modeled in mice. These studies emphasize caution in the genetic diagnoses of infertile patients based primarily on pathogenicity prediction algorithms and emphasize the need for alternative and efficient in vitro or in vivo functional validation models for more effective and accurate VUS description to either pathogenic or benign categories.

Implementation of machine learning models to determine the appropriate model for protein function prediction

Predicting the function of proteins is a crucial part of genome annotation, which can help in solving a wide range of biological problems. Many methods are available to predict the functions of proteins. However, except for sequence, most features are difficult to obtain or are not available for many proteins, which limits their scope. In addition, the performance of sequence-based feature prediction methods is often lower than that of methods that involve multiple features, and protein feature prediction can be time-consuming. Recent advances in this field are associated with the development of machine learning, which shows great progress in solving the problem of predicting protein functions. Today, however, most protein sequences have the status of «uncharacterized» or «putative». The need to assess the accuracy of identification of protein functions is an urgent task for machine learning approaches used to predict protein functions. In this study, the performance of two popular function prediction algorithms (ProtCNN and BiLSTM) was assessed from two perspectives and the procedures for building these models were described. As a result of the study of Pfam families, ProtCNN achieves an accuracy rate of 0.988 % and bidirectional LSTM has an accuracy rate of 0.9506 %. The use of the Pfam dataset allowed increasing the classification accuracy due to the large training dataset. The quality of the prediction increases with a large amount of training data. The study demonstrated that machine learning algorithms can be used as an effective tool for building protein function prediction models, in particular, the CNN network can be adapted as an accurate tool for annotating protein functions in the presence of large datasets.

Resistant starch type-4 intake alters circulating bile acids in human subjects.

Resistant starch (RS) type 4 (RS4) is a type of RS, a class of non-digestible prebiotic dietary fibers with a range of demonstrated metabolic health benefits to the host. On the other hand, bile acids (BA) have recently emerged as an important class of metabolic function mediators that involve host-microbiota interactions. RS consumption alters fecal and cecal BA in humans and rodents, respectively. The effect of RS intake on circulating BA concentrations remains unexplored in humans. Using available plasma and stool samples from our previously reported double-blind, controlled, 2-arm crossover nutrition intervention trial (Clinicaltrials.gov: NCT01887964), a liquid-chromatography/mass-spectrometry-based targeted multiple reaction monitoring, and absolute quantifications, we assessed BA changes after 12 weeks of an average 12 g/day RS4-intake. Stool BA concentrations were lower post RS4 compared to the control, the two groups consuming similar macronutrients (n = 14/group). Partial least squares-discriminant analysis revealed distinct BA signatures in stool and plasma post interventions. The increased circulating BA concentrations were further investigated using linear mixed-effect modeling that controlled for potential confounders. A higher plasma abundance of several BA species post RS4 was observed (fold increase compared to control in parenthesis): taurocholic acid (1.92), taurodeoxycholic acid (1.60), glycochenodeoxycholic acid (1.58), glycodeoxycholic acid (1.79), and deoxycholic acid (1.77) (all, p < 0.05). Distinct microbiome ortholog-signatures were observed between RS4 and control groups (95% CI), derived using the Piphillin function-prediction algorithm and principal component analysis (PCA) of pre-existing 16S rRNA gene sequences. Association of Bifidobacterium adolescentis with secondary BA such as, deoxycholic acid (rho = 0.55, p = 0.05), glycodeoxycholic acid (rho = 0.65, p = 0.02), and taurodeoxycholic acid (rho = 0.56, p = 0.04) were observed in the RS4-group, but not in the control group (all, p > 0.05). Our observations indicate a previously unknown in humans- RS4-associated systemic alteration of microbiota-derived secondary BA. Follow-up investigations of BA biosynthesis in the context of RS4 may provide molecular targets to understand and manipulate microbiome-host interactions.

Clinical Practice of Targeted Capture Sequencing to Identify Actionable Alterations in Cholangiocarcinoma.

Simple SummaryGenetic testing promises to provide guidance for early diagnosis and treatment of cholangiocarcinoma (CCA). Due to the different mutation landscapes across populations and the paucity of sequencing data of Chinese patients with CCA, the existing mutation landscape is insufficient to reflect the mutation characteristics of Chinese patients. We retrospectively analyzed 72 Chinese patients with CCA who had received genetic testing of targeted capture sequencing. A total of 2152 somatic mutations were detected in 56 (77.78%, 56/72) patients, of which, the frequently mutated driver genes were TP53, KMT2D, KMT2C, BCOR, APC, BAP1, ARID1A, NF1, PIK3CA, KRAS, and LRP1B. Most mutations were enriched in NRF2, TP53, and TGF-Beta oncogenic signaling pathways and cadherin repeat domain which were associated with intercellular adhesion. We identified 118 novel pathogenic or likely pathogenic somatic mutations and 77 actionable alterations, which provided potential targets for early diagnosis and treatment planning of CCA.The early diagnosis and treatment of cholangiocarcinoma (CCA) remain a challenge worldwide. Genetic testing promises to solve these problems. Due to the different mutation landscapes across populations and the paucity of sequencing data of Chinese patients with CCA, the existing mutation landscape is insufficient to reflect the mutation characteristics of Chinese patients. Thus, we retrospectively analyzed 72 Chinese patients with CCA who had received genetic testing of targeted capture sequencing. A total of 2152 somatic mutations were detected in 56 (77.78%) patients, of which, the frequently mutated driver genes were TP53 (27.78%), KMT2D (23.81%), KMT2C (20.63%), BCOR (18.06%), APC (15.28%), BAP1 (13.89%), ARID1A (12.50%), NF1 (12.50%), PIK3CA (12.50%), KRAS (11.11%), and LRP1B (11.11%). Most mutations were enriched in NRF2, TP53, and TGF-Beta oncogenic signaling pathways and cadherin repeat domains which were associated with intercellular adhesion. Based on cancer-related public databases and multiple protein function prediction algorithms, we identified 118 novel pathogenic or likely pathogenic somatic mutations and 77 actionable alterations. Molecular analysis of tumors from a precision oncology perspective can provide potential targets for early diagnosis and treatment of CCA and assist physicians in clinical decision making.

Organization of gene programs revealed by unsupervised analysis of diverse gene-trait associations.

Genome wide association studies provide statistical measures of gene–trait associations that reveal how genetic variation influences phenotypes. This study develops an unsupervised dimensionality reduction method called UnTANGLeD (Unsupervised Trait Analysis of Networks from Gene Level Data) which organizes 16,849 genes into discrete gene programs by measuring the statistical association between genetic variants and 1,393 diverse complex traits. UnTANGLeD reveals 173 gene clusters enriched for protein–protein interactions and highly distinct biological processes governing development, signalling, disease, and homeostasis. We identify diverse gene networks with robust interactions but not associated with known biological processes. Analysis of independent disease traits shows that UnTANGLeD gene clusters are conserved across all complex traits, providing a simple and powerful framework to predict novel gene candidates and programs influencing orthogonal disease phenotypes. Collectively, this study demonstrates that gene programs co-ordinately orchestrating cell functions can be identified without reliance on prior knowledge, providing a method for use in functional annotation, hypothesis generation, machine learning and prediction algorithms, and the interpretation of diverse genomic data.

In silico analyses of acetylcholinesterase (AChE) and its genetic variants in interaction with the anti-Alzheimer drug Rivastigmine.

Alzheimer's disease (AD) is the leading cause of dementia worldwide. Despite causing great social and economic impact, there is currently no cure for AD. The most effective therapy to manage AD symptoms is based on acetylcholinesterase inhibitors (AChEi), from which rivastigmine presented numerous benefits. However, mutations in AChE, which affect approximately 5% of the population, can modify protein structure and function, changing the individual response to Alzheimer's treatment. In this study, we performed computer simulations of AChE wild type and variants R34Q, P135A, V333E, and H353N, identified by one or more genome-wide association studies, to evaluate their effects on protein structure and interaction with rivastigmine. The functional effects of AChE variants were predicted using eight machine learning algorithms, while the evolutionary conservation of AChE residues was analyzed using the ConSurf server. Autodock4.2.6 was used to predict the binding modes for the hAChE-rivastigmine complex, which is still unknown. Molecular dynamics (MD) simulations were performed in triplicates for the AChE wild type and mutants using the GROMACS packages. Among the analyzed variants, P135A was classified as deleterious by all the functional prediction algorithms, in addition to occurring at highly conserved positions, which may have harmful consequences on protein function. The molecular docking results suggested that rivastigmine interacts with hAChE at the upper active-site gorge, which was further confirmed by MD simulations. Our MD findings also suggested that the complex hAChE-rivastigmine remains stable over time. The essential dynamics revealed flexibility alterations at the active-site gorge upon mutations P135A, V333E, and H353N, which may lead to strong and nonintuitive consequences to hAChE binding. Nonetheless, similar binding affinities were registered in the MMPBSA analysis for the hAChE wild type and variants when complexed to rivastigmine. Finally, our findings indicated that the rivastigmine binding to hAChE is an energetically favorable process mainly driven by negatively charged amino acids.

NKX2-5 Variant in Two Siblings with Thyroid Hemiagenesis.

Thyroid hemiagenesis (THA) is an inborn absence of one thyroid lobe of largely unknown etiopathogenesis. The aim of the study was to reveal genetic factors responsible for thyroid maldevelopment in two siblings with THA. None of the family members presented with congenital heart defect. The samples were subjected to whole-exome sequencing (WES) (Illumina, TruSeq Exome Enrichment Kit, San Diego, CA 92121, USA). An ultra-rare variant c.839C>T (p.Pro280Leu) in NKX2-5 gene (NM_004387.4) was identified in both affected children and an unaffected father. In the mother, the variant was not present. This variant is reported in population databases with 0.0000655 MAF (GnomAD v3, dbSNP rs761596254). The affected amino acid position is moderately conserved (positive scores in PhyloP: 1.364 and phastCons: 0.398). Functional prediction algorithms showed deleterious impact (dbNSFP v4.1, FATHMM, SIFT) or benign (CADD, PolyPhen-2, Mutation Assessor). According to ACMG criteria, variant is classified as having uncertain clinical significance. For the first time, NKX2-5 gene variants were found in two siblings with THA, providing evidence for its potential contribution to the pathogenesis of this type of thyroid dysgenesis. The presence of the variant in an unaffected parent, carrier of p.Pro280Leu variant, suggests potential contribution of yet unidentified additional factors determining the final penetrance and expression.

Identifying Actionable Variants Using Capture-Based Targeted Sequencing in 563 Patients With Non-Small Cell Lung Carcinoma.

Although the NSCLC diagnostic standards recommend the detection of driver gene mutation, comprehensive genomic profiling has not been used widely in clinical practice. As to the different mutation spectrum characteristics between populations, the research based on Chinese NSCLC cohort is very important for clinical practice. Therefore, we collected 563 surgical specimens from patients with non-small cell lung carcinoma and applied capture-based sequencing using eight-gene panel. We identified 556 variants, with 416 potentially actionable variants in 54.88% (309/563) patients. These single nucleotide variants, insertions and deletions were most commonly found in EGFR (55%), followed by ERBB2 (12%), KRAS (11%), PIK3CA (9%), MET (8%), BRAF (7%), DDR2 (2%), NRAS (0.3%). By using ten protein function prediction algorithms, we also identified 30 novel potentially pathogenic variants. Ninety-eight patients harbored EFGR exon 21 p.L858R mutation and the catalytic domain of the protein tyrosine kinase (PTKc) in EGFR is largely mutated. In addition, there were nine frequent pathogenic variants found in five or more patients. This data provides the potential molecular basis for directing the treatment of lung cancer.

High-pressure transformations of CaC2O5 - a full structural trend from double [CO3] triangles through the isolated group of [CO4] tetrahedra to framework and layered structures.

Over the past few years, the concept of carbonates, as the salts of MCO3 or composition with [CO3] triangles in the crystal structures, was sufficiently extended. In addition to carbonates, crystal structures with stoichiometry M3CO5, M2CO4 and MC2O5 were predicted and successfully synthesized. In the present study, based on density functional theory and crystal structure prediction algorithms, we found a novel structure of CaC2O5, namely Ca-pyrocarbonate with monoclinic symmetry Cc, which is one of the possible agents of the global carbon cycle. This structure is characterized by the isolated [C2O5] groups consisting of two [CO3] triangles connected through a common oxygen atom. The thermodynamic stability field of Ca-pyrocarbonate with respect to the decomposition reaction into calcium carbonate and carbon dioxide begins at a pressure of 10 GPa. As the pressure increases to 21 GPa, the structure of Ca-pyrocarbonate transforms into the recently synthesized tetragonal modification I4̄2d, in the structure of which carbon is in the sp3-hybridized state and [CO4] tetrahedra form isolated pyramidal [C4O10] anionic groups. At 59 GPa in the temperature range of 0-2500 K, CaC2O5-I4̄2d undergoes a phase transition to CaC2O5-Fdd2, with the framework structure of [CO4] tetrahedra. On further compression to about 80 GPa, the framework structure transforms into layered ones, C2 and Pc. In addition, we estimated the thermodynamic stability of CaC2O5 with respect to the minerals of the Earth's mantle. We found that CaC2O5 can coexist with bridgmanite up to pressures of 54 GPa at 300 K, where it reacts with the formation of a Ca-perovskite, magnesite, and solid CO2-V.

Alkali Metal (Li, Na, and K) Orthocarbonates: Stabilization of sp3-Bonded Carbon at Pressures above 20 GPa

In the present work, based on the density functional theory and crystal structure prediction algorithms, we found novel structures of alkali metal orthocarbonates Li4CO4-C2/c, Na4CO4-C2/c, and K4CO4-P21/c. The pressure of thermodynamic stabilization of the obtained orthocarbonates with respect to decomposition into the corresponding carbonate and oxide ranges from 21 GPa (for K4CO4) to 25 GPa (for Li4CO4). All obtained structures contain isolated groups of [CO4]tetrahedra. The structures of lithium and sodium orthocarbonates were found to be isotypic. As well as alkaline earth orthocarbonates, orthocarbonates of alkali metals are structurally similar to orthosilicates of the same metal. Na4CO4-C2/c was found to be dynamically stable at ambient pressure, while all other structures are destabilized on decompression. In addition, the P–T phase diagrams of lithium, sodium, and potassium oxides have been calculated for the first time. All oxides have a high-pressure phase transition from the Fm3̅m (anti-fluorite) structure to the Pnma (anti-cotunnite) structure at 34 GPa for Li2O, at 10 GPa for Na2O, and at 5 GPa for K2O at low temperatures. In the Na2O and K2O systems, high-temperature phase transition to the P6̅2m (barringerite) structure was revealed at temperatures above 1300 and 650 K, respectively.

Cover Image, Volume 89, Issue 11

Front Cover: The cover image is based on the Original Article A fast protein binding site comparison algorithm for proteome-wide protein function prediction and drug repurposing by Honglin Li et al., https://doi.org/10.1002/prot.26176

Fe–N System at High Pressures and Its Relevance to the Earth’s Core Composition

Based on ab initio calculations within the density functional theory and crystal structure prediction algorithms, the structure and stability of iron–nitrogen compounds in the pressure range of 100...

Phase Relations in the Ni–S System at High Pressures from ab Initio Computations

Based on the ab initio calculations within the density functional theory and crystal structure prediction algorithms, the structure and stability of compounds in the Ni–S system at pressures of 100–400 GPa were determined. As a result, a homologous series of discrete compounds (Ni and S) consisting of Ni14S-C2/m, Ni13S-R3̅, Ni12S-R3̅, Ni5S-C2/m, Ni4S-P1̅, and Ni3S-Cmcm is revealed. We also confirmed the absence of the stable Fe-bearing compounds between Fe and Fe2S in the studied pressure range. At the Earth’s core pressures, 4 wt % of sulfur can be dissolved in solid fcc-Ni without deformation of the structure. Significant deformations in the Ni structure occur at sulfur contents from 4 to 15 wt %. In contrast, up to 0.45 wt % of sulfur could be dissolved in hcp-Fe at 350 GPa and 0 K. For Ni3S, two phases with space groups I4̅ and Cmcm were predicted. Ni3S-I4̅ is stable at least from 100 GPa, whereas above 330 GPa, it transforms into Ni3S-Cmcm. The pressure of phase transition is almost independent of temperature. The Ni2S is stable in the entire pressure range and undergoes a single-phase transition from the Pnma- to P6̅2m-phase at 266 GPa and 0 K with a Clapeyron slope of 5 MPa/K. The S-rich sulfide NiS3 is characterized by Im3̅m symmetry and is thermodynamically stable from 100 to 318 GPa. Our new data on Ni sulfides might be important to constrain detailed thermodynamic models for Fe–Ni-bearing Earth and planetary cores.

DeepCys: Structure-based multiple cysteine function prediction method trained on deep neural network: Case study on domains of unknown functions belonging to COX2 domains.

Cysteine (Cys) is the most reactive amino acid participating in a wide range of biological functions. In-silico predictions complement the experiments to meet the need of functional characterization. Multiple Cys function prediction algorithm is scarce, in contrast to specific function prediction algorithms. Here we present a deep neural network-based multiple Cys function prediction, available on web-server (DeepCys) (https://deepcys.herokuapp.com/). DeepCys model was trained and tested on two independent datasets curated from protein crystal structures. This prediction method requires three inputs, namely, PDB identifier (ID), chain ID and residue ID for a given Cys and outputs the probabilities of four cysteine functions, namely, disulphide, metal-binding, thioether and sulphenylation and predicts the most probable Cys function. The algorithm exploits the local and global protein properties, like, sequence and secondary structure motifs, buried fractions, microenvironments and protein/enzyme class. DeepCys outperformed most of the multiple and specific Cys function algorithms. This method can predict maximum number of cysteine functions. Moreover, for the first time, explicitly predicts thioether function. This tool was used to elucidate the cysteine functions on domains of unknown functions belonging to cytochrome C oxidase subunit-II like transmembrane domains. Apart from the web-server, a standalone program is also available on GitHub (https://github.com/vam-sin/deepcys).

From miRNA Target Gene Network to miRNA Function: miR-375 Might Regulate Apoptosis and Actin Dynamics in the Heart Muscle via Rho-GTPases-Dependent Pathways.

MicroRNAs (miRNAs) are short, single-stranded, non-coding ribonucleic acid (RNA) molecules, which are involved in the regulation of main biological processes, such as apoptosis or cell proliferation and differentiation, through sequence-specific interaction with target mRNAs. In this study, we propose a workflow for predicting miRNAs function by analyzing the structure of the network of their target genes. This workflow was applied to study the functional role of miR-375 in the heart muscle (myocardium), since this miRNA was previously shown to be associated with heart diseases, and data on its function in the myocardium are mostly unclear. We identified PIK3CA, RHOA, MAPK3, PAFAH1B1, CTNNB1, MYC, PRKCA, ERBB2, and CDC42 as key genes in the miR-375 regulated network and predicted the possible function of miR-375 in the heart muscle, consisting mainly in the regulation of the Rho-GTPases-dependent signaling pathways. We implemented our algorithm for miRNA function prediction into a Python module, which is available at GitHub.

Predicting Protein Functions Based on Differential Co-expression and Neighborhood Analysis.

Proteins are polypeptides essential in biological processes. Protein physical interactions are complemented by other types of functional relationship data including genetic interactions, knowledge about co-expression, and evolutionary pathways. Existing algorithms integrate protein interaction and gene expression data to retrieve context-specific subnetworks composed of genes/proteins with known and unknown functions. However, most protein function prediction algorithms fail to exploit diverse intrinsic information in feature and label spaces. We develop a novel integrative method based on differential Co-expression analysis and Neighbor-voting algorithm for Protein Function Prediction, namely CNPFP. The method integrates heterogeneous data and exploits intrinsic and latent linkages via global iterative approach and genomic features. CNPFP performs three tasks: clustering, differential co-expression analysis, and predicts protein functions. Our aim is to identify yeast cell cycle-specific proteins linked to differentially expressed proteins in the protein-protein interaction network. To capture intrinsic information, CNPFP selects the most relevant feature subset based on global iterative neighbor-voting algorithm. We identify eight condition-specific modules. The most relevant subnetwork has 87 genes highly enriched with cyclin-dependent kinases, a protein kinase relevant for cell cycle regulation. We present comprehensive annotations for 3538 Saccharomyces cerevisiae proteins. Our method achieves an AUROC of 0.9862, accuracy of 0.9710, and F-score of 0.9691. From the results, we can summarize that exploiting intrinsic nature of protein relationships improves the quality of function prediction. Thus, the proposed method is useful in functional genomics studies.

Phase relations in the Fe-P system at high pressures and temperatures from ab initio computations

ABSTRACTBased on the first-principles calculations within the density functional theory and crystal structure prediction algorithms iron phosphide phases stable under pressure of the Earth’s core and temperatures up to 4000 K were determined. A new low-temperature modification FeP-P21/c stable above ∼75 GPa was predicted. Fe2P with the allabogdanite structure has been established to be stable in the low-temperature region at ambient conditions. At 750 K it transforms into the barringerite structure. The transition from Fe3P with schreibersite structure to Fe3P-Cmcm was observed at 27 GPa, and the phase transition boundary is nearly isobaric. Fe2P and FeP are thermodynamically stable at the Earth’s inner core pressures and 0 K according to the obtained results, whereas Fe3P stabilizes with respect to decomposition to Fe + Fe2P at high temperatures above ∼3200 K.

MTRR gene variant rs1801394 found in Malaysian patients with neural tube defects

Neural tube defects (NTDs) are congenital anomalies resulting from the failure of neural tube closure during embryogenesis. The precise molecular mechanisms underlying this multifactorial disease is poorly understood, although single nucleotide polymorphisms in genes involved in the one-carbon metabolism cycle are believed to contribute towards NTD development. Among them is 5-methyltetrahydrofolate-homocysteine methyltransferase reductase (MTRR). Protein function prediction algorithms (PolyPhen-2, PROVEAN, SIFT, SMART-Ensembl) were employed to evaluate its pathogenicity potential caused by the replacement of isoleucine with methionine. Seven NTD patients and 12 of their parents were recruited for this study. DNA samples were collected through blood or saliva whereby the extracted DNAs were then sent for whole exome sequencing (WES). Zygosity of the variant was confirmed from WES data of each subject and further validated through polymerase chain reaction (PCR) and Sanger sequencing. The results revealed that 57% of patients and 83% of parents carried rs1801394 mutation in their MTRR gene, based on either homozygous (G/G) or heterozygous (A/G) genotypes. Bioinformatics analysis of this missense mutation predicted that this change is damaging to MTRR protein function by 2 of the 3 predictor algorithms and that the change from isoleucine to methionine amino acid affects flavodoxin domain of the protein. This impacts enzyme activity within the one-carbon metabolism pathway, which is linked to the aetiology of NTDs. From population databases, this variant was considered common with a MAF &gt;0.3, however, it was not found in the Singapore Genome Variation Project (SGVP), whose population is a closer representation of the Malaysian subjects investigated here. Hence, we explored the prevalence of this variant in other studies and found that its association with NTDs differed across populations worldwide. Finally, we conclude that rs1801394 may be an NTD risk factor in the Malaysian population and should be further investigated as a potential prenatal screening tool.