SSpro/ACCpro 6: almost perfect prediction of protein secondary structure and relative solvent accessibility using profiles, deep learning and structural similarity.

MOTIVATION: Accurately predicting protein secondary structure and relative solvent accessibility is important for the study of protein evolution, structure and an early-stage component of typical protein 3D structure prediction pipelines. RESULTS: We present a new improved version of the SSpro/ACCpro suite of predictors for the prediction of protein secondary structure (in three and eight classes) and relative solvent accessibility. The changes include improved, TensorFlow-trained, deep learning predictors, a richer set of profile features (232 features per residue position) and sequence-only features (71 features per position), a more recent Protein Data Bank (PDB) snapshot for training, better hyperparameter tuning and improvements made to the HOMOLpro module, which leverages structural information from protein segment homologs in the PDB. The new SSpro 6 outperforms the previous version (SSpro 5) by 3-4% in Q3 accuracy and, when used with HOMOLPRO, reaches accuracy in the 95-100% range. AVAILABILITY AND IMPLEMENTATION: The predictors' software, data and web servers are available through the SCRATCH suite of protein structure predictors at http://scratch.proteomics.ics.uci.edu. To maximize comptatibility and ease of use, the deep learning predictors are re-implemented as pure Python/numpy code without TensorFlow dependency. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Protein profiles: Biases and protocols.

The use of evolutionary profiles to predict protein secondary structure, as well as other protein structural features, has been standard practice since the 1990s. Using profiles in the input of such predictors, in place or in addition to the sequence itself, leads to significantly more accurate predictions. While profiles can enhance structural signals, their role remains somewhat surprising as proteins do not use profiles when folding in vivo. Furthermore, the same sequence-based redundancy reduction protocols initially derived to train and evaluate sequence-based predictors, have been applied to train and evaluate profile-based predictors. This can lead to unfair comparisons since profiles may facilitate the bleeding of information between training and test sets. Here we use the extensively studied problem of secondary structure prediction to better evaluate the role of profiles and show that: (1) high levels of profile similarity between training and test proteins are observed when using standard sequence-based redundancy protocols; (2) the gain in accuracy for profile-based predictors, over sequence-based predictors, strongly relies on these high levels of profile similarity between training and test proteins; and (3) the overall accuracy of a profile-based predictor on a given protein dataset provides a biased measure when trying to estimate the actual accuracy of the predictor, or when comparing it to other predictors. We show, however, that this bias can be mitigated by implementing a new protocol (EVALpro) which evaluates the accuracy of profile-based predictors as a function of the profile similarity between training and test proteins. Such a protocol not only allows for a fair comparison of the predictors on equally hard or easy examples, but also reduces the impact of choosing a given similarity cutoff when selecting test proteins. The EVALpro program is available in the SCRATCH suite ( www.scratch.proteomics.ics.uci.edu) and can be downloaded at: www.download.igb.uci.edu/#evalpro.

Defining the Independence of the Liver Circadian Clock.

Mammals rely on a network of circadian clocks to control daily systemic metabolism and physiology. The central pacemaker in the suprachiasmatic nucleus (SCN) is considered hierarchically dominant over peripheral clocks, whose degree of independence, or tissue-level autonomy, has never been ascertained in vivo. Using arrhythmic Bmal1-null mice, we generated animals with reconstituted circadian expression of BMAL1 exclusively in the liver (Liver-RE). High-throughput transcriptomics and metabolomics show that the liver has independent circadian functions specific for metabolic processes such as the NAD+ salvage pathway and glycogen turnover. However, although BMAL1 occupies chromatin at most genomic targets in Liver-RE mice, circadian expression is restricted to ∼10% of normally rhythmic transcripts. Finally, rhythmic clock gene expression is lost in Liver-RE mice under constant darkness. Hence, full circadian function in the liver depends on signals emanating from other clocks, and light contributes to tissue-autonomous clock function.

Epigenetic regulation of the circadian gene Per1 contributes to age-related changes in hippocampal memory.

Aging is accompanied by impairments in both circadian rhythmicity and long-term memory. Although it is clear that memory performance is affected by circadian cycling, it is unknown whether age-related disruption of the circadian clock causes impaired hippocampal memory. Here, we show that the repressive histone deacetylase HDAC3 restricts long-term memory, synaptic plasticity, and experience-induced expression of the circadian gene Per1 in the aging hippocampus without affecting rhythmic circadian activity patterns. We also demonstrate that hippocampal Per1 is critical for long-term memory formation. Together, our data challenge the traditional idea that alterations in the core circadian clock drive circadian-related changes in memory formation and instead argue for a more autonomous role for circadian clock gene function in hippocampal cells to gate the likelihood of long-term memory formation.

Mutation of neuron-specific chromatin remodeling subunit BAF53b: rescue of plasticity and memory by manipulating actin remodeling.

Recent human exome-sequencing studies have implicated polymorphic Brg1-associated factor (BAF) complexes (mammalian SWI/SNF chromatin remodeling complexes) in several intellectual disabilities and cognitive disorders, including autism. However, it remains unclear how mutations in BAF complexes result in impaired cognitive function. Post-mitotic neurons express a neuron-specific assembly, nBAF, characterized by the neuron-specific subunit BAF53b. Subdomain 2 of BAF53b is essential for the differentiation of neuronal precursor cells into neurons. We generated transgenic mice lacking subdomain 2 of Baf53b (BAF53bΔSB2). Long-term synaptic potentiation (LTP) and long-term memory, both of which are associated with phosphorylation of the actin severing protein cofilin, were assessed in these animals. A phosphorylation mimic of cofilin was stereotaxically delivered into the hippocampus of BAF53bΔSB2 mice in an effort to rescue LTP and memory. BAF53bΔSB2 mutant mice show impairments in phosphorylation of synaptic cofilin, LTP, and memory. Both the synaptic plasticity and memory deficits are rescued by overexpression of a phosphorylation mimetic of cofilin. Baseline physiology and behavior were not affected by the mutation or the experimental treatment. This study suggests a potential link between nBAF function, actin cytoskeletal remodeling at the dendritic spine, and memory formation. This work shows that a targeted manipulation of synaptic function can rescue adult plasticity and memory deficits caused by manipulations of nBAF, and thereby provides potential novel avenues for therapeutic development for multiple intellectual disability disorders.

VIRALpro: a tool to identify viral capsid and tail sequences.

MOTIVATION: Not only sequence data continue to outpace annotation information, but also the problem is further exacerbated when organisms are underrepresented in the annotation databases. This is the case with non-human-pathogenic viruses which occur frequently in metagenomic projects. Thus, there is a need for tools capable of detecting and classifying viral sequences. RESULTS: We describe VIRALpro a new effective tool for identifying capsid and tail protein sequences, which are the cornerstones toward viral sequence annotation and viral genome classification. AVAILABILITY AND IMPLEMENTATION: The data, software and corresponding web server are available from http://scratch.proteomics.ics.uci.edu as part of the SCRATCH suite. CONTACT: clovis.galiez@inria.fr or pfbaldi@uci.edu SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SSpro/ACCpro 5: almost perfect prediction of protein secondary structure and relative solvent accessibility using profiles, machine learning and structural similarity.

MOTIVATION: Accurately predicting protein secondary structure and relative solvent accessibility is important for the study of protein evolution, structure and function and as a component of protein 3D structure prediction pipelines. Most predictors use a combination of machine learning and profiles, and thus must be retrained and assessed periodically as the number of available protein sequences and structures continues to grow. RESULTS: We present newly trained modular versions of the SSpro and ACCpro predictors of secondary structure and relative solvent accessibility together with their multi-class variants SSpro8 and ACCpro20. We introduce a sharp distinction between the use of sequence similarity alone, typically in the form of sequence profiles at the input level, and the additional use of sequence-based structural similarity, which uses similarity to sequences in the Protein Data Bank to infer annotations at the output level, and study their relative contributions to modern predictors. Using sequence similarity alone, SSpro's accuracy is between 79 and 80% (79% for ACCpro) and no other predictor seems to exceed 82%. However, when sequence-based structural similarity is added, the accuracy of SSpro rises to 92.9% (90% for ACCpro). Thus, by combining both approaches, these problems appear now to be essentially solved, as an accuracy of 100% cannot be expected for several well-known reasons. These results point also to several open technical challenges, including (i) achieving on the order of ≥ 80% accuracy, without using any similarity with known proteins and (ii) achieving on the order of ≥ 85% accuracy, using sequence similarity alone. AVAILABILITY AND IMPLEMENTATION: SSpro, SSpro8, ACCpro and ACCpro20 programs, data and web servers are available through the SCRATCH suite of protein structure predictors at http://scratch.proteomics.ics.uci.edu.

A Genomic Analysis Pipeline and Its Application to Pediatric Cancers.

We present a cancer genomic analysis pipeline which takes as input sequencing reads for both germline and tumor genomes and outputs filtered lists of all genetic mutations in the form of short ranked list of the most affected genes in the tumor, using either the Complete Genomics or Illumina platforms. A novel reporting and ranking system has been developed that makes use of publicly available datasets and literature specific to each patient, including new methods for using publicly available expression data in the absence of proper control data. Previously implicated small and large variations (including gene fusions) are reported in addition to probable driver mutations. Relationships between cancer and the sequenced tumor genome are highlighted using a network-based approach that integrates known and predicted protein-protein, protein-TF, and protein-drug interaction data. By using an integrative approach, effects of genetic variations on gene expression are used to provide further evidence of driver mutations. This pipeline has been developed with the aim to be used in assisting in the analysis of pediatric tumors, as an unbiased and automated method for interpreting sequencing results along with identifying potentially therapeutic drugs and their targets. We present results that agree with previous literature and highlight specific findings in a few patients.

The neuron-specific chromatin regulatory subunit BAF53b is necessary for synaptic plasticity and memory.

Recent exome sequencing studies have implicated polymorphic Brg1-associated factor (BAF) complexes (mammalian SWI/SNF chromatin remodeling complexes) in several human intellectual disabilities and cognitive disorders. However, it is currently unknown how mutations in BAF complexes result in impaired cognitive function. Postmitotic neurons express a neuron-specific assembly, nBAF, characterized by the neuron-specific subunit BAF53b. Mice harboring selective genetic manipulations of BAF53b have severe defects in long-term memory and long-lasting forms of hippocampal synaptic plasticity. We rescued memory impairments in BAF53b mutant mice by reintroducing BAF53b in the adult hippocampus, which suggests a role for BAF53b beyond neuronal development. The defects in BAF53b mutant mice appeared to derive from alterations in gene expression that produce abnormal postsynaptic components, such as spine structure and function, and ultimately lead to deficits in synaptic plasticity. Our results provide new insight into the role of dominant mutations in subunits of BAF complexes in human intellectual and cognitive disorders.

Mitochondrial mutations and polymorphisms in psychiatric disorders.

Mitochondrial deficiencies with unknown causes have been observed in schizophrenia (SZ) and bipolar disorder (BD) in imaging and postmortem studies. Polymorphisms and somatic mutations in mitochondrial DNA (mtDNA) were investigated as potential causes with next generation sequencing of mtDNA (mtDNA-Seq) and genotyping arrays in subjects with SZ, BD, major depressive disorder (MDD), and controls. The common deletion of 4,977 bp in mtDNA was compared between SZ and controls in 11 different vulnerable brain regions and in blood samples, and in dorsolateral prefrontal cortex (DLPFC) of BD, SZ, and controls. In a separate analysis, association of mitochondria SNPs (mtSNPs) with SZ and BD in European ancestry individuals (n = 6,040) was tested using Genetic Association Information Network (GAIN) and Wellcome Trust Case Control Consortium 2 (WTCCC2) datasets. The common deletion levels were highly variable across brain regions, with a 40-fold increase in some regions (nucleus accumbens, caudate nucleus and amygdala), increased with age, and showed little change in blood samples from the same subjects. The common deletion levels were increased in the DLPFC for BD compared to controls, but not in SZ. Full mtDNA genome resequencing of 23 subjects, showed seven novel homoplasmic mutations, five were novel synonymous coding mutations. By logistic regression analysis there were no significant mtSNPs associated with BD or SZ after genome wide correction. However, nominal association of mtSNPs (p < 0.05) to SZ and BD were found in the hypervariable region of mtDNA to T195C and T16519C. The results confirm prior reports that certain brain regions accumulate somatic mutations at higher levels than blood. The study in mtDNA of common polymorphisms, somatic mutations, and rare mutations in larger populations may lead to a better understanding of the pathophysiology of psychiatric disorders.

High-throughput prediction of protein antigenicity using protein microarray data.

MOTIVATION: Discovery of novel protective antigens is fundamental to the development of vaccines for existing and emerging pathogens. Most computational methods for predicting protein antigenicity rely directly on homology with previously characterized protective antigens; however, homology-based methods will fail to discover truly novel protective antigens. Thus, there is a significant need for homology-free methods capable of screening entire proteomes for the antigens most likely to generate a protective humoral immune response. RESULTS: Here we begin by curating two types of positive data: (i) antigens that elicit a strong antibody response in protected individuals but not in unprotected individuals, using human immunoglobulin reactivity data obtained from protein microarray analyses; and (ii) known protective antigens from the literature. The resulting datasets are used to train a sequence-based prediction model, ANTIGENpro, to predict the likelihood that a protein is a protective antigen. ANTIGENpro correctly classifies 82% of the known protective antigens when trained using only the protein microarray datasets. The accuracy on the combined dataset is estimated at 76% by cross-validation experiments. Finally, ANTIGENpro performs well when evaluated on an external pathogen proteome for which protein microarray data were obtained after the initial development of ANTIGENpro. AVAILABILITY: ANTIGENpro is integrated in the SCRATCH suite of predictors available at http://scratch.proteomics.ics.uci.edu. CONTACT: pfbaldi@ics.uci.edu

SOLpro: accurate sequence-based prediction of protein solubility.

MOTIVATION: Protein insolubility is a major obstacle for many experimental studies. A sequence-based prediction method able to accurately predict the propensity of a protein to be soluble on overexpression could be used, for instance, to prioritize targets in large-scale proteomics projects and to identify mutations likely to increase the solubility of insoluble proteins. RESULTS: Here, we first curate a large, non-redundant and balanced training set of more than 17 000 proteins. Next, we extract and study 23 groups of features computed directly or predicted (e.g. secondary structure) from the primary sequence. The data and the features are used to train a two-stage support vector machine (SVM) architecture. The resulting predictor, SOLpro, is compared directly with existing methods and shows significant improvement according to standard evaluation metrics, with an overall accuracy of over 74% estimated using multiple runs of 10-fold cross-validation.