MOASL: Predicting drug mechanism of actions through similarity learning with transcriptomic signature.

Understanding the mechanisms of actions (MOAs) of compounds is crucial in drug discovery. A common step in drug MOAs annotation is to query the dysregulated gene signatures induced by drugs in a reference library of pre-defined signatures. However, traditional similarity-based computational strategies face challenges when dealing with high-dimensional and noisy transcriptional signature data. To address this issue, we introduce MOASL (MOAs prediction via Similarity Learning), a novel approach that contrastive to learn similarity embeddings among signatures with shared MOAs automatically. We evaluated the accuracy of signature matching on various transcriptional activity score (TAS) datasets and individual cell lines by using MOASL. The results show MOASL achieved higher performance over several statistical and machine learning methods. Furthermore, we provided the rationale of our model by visualizing the signature annotation procedure. Using MOASL, the MOAs label of query signature could be conveniently defined by calculating the similarity between the query embedding and the reference embeddings. Finally, we applied MOASL to repurpose thousands of compounds as glucocorticoid receptor (GR) agonists, accurately identifying 8 out of the top 10 compounds. MOASL is conveniently accessible on GitHub at https://github.com/jianglikun/MOASL, empowering researchers and practitioners in the field of drug discovery to predict the MOAs of drug.

The effects of methylphenidate and atomoxetine on Drosophila brain at single-cell resolution and potential drug repurposing for ADHD treatment.

The stimulant methylphenidate (MPH) and the non-stimulant atomoxetine (ATX) are frequently used for the treatment of attention-deficit/hyperactivity disorder (ADHD); however, the function of these drugs in different types of brain cells and their effects on related genes remain largely unknown. To address these questions, we built a pipeline for the simultaneous examination of the activity behavior and transcriptional responses of Drosophila melanogaster at single-cell resolution following drug treatment. We selected the Drosophila with significantly increased locomotor activities (hyperactivity-like behavior) following the administration of each drug in comparison with the control (same food as the drug-treated groups with 5% sucrose, yeast, and blue food dye solution) using EasyFlyTracker. Subsequently, single cell RNA sequencing (scRNASEQ) was used to capture the transcriptome of 82,917 cells, unsupervised clustering analysis of which yielded 28 primary cell clusters representing the major cell types in adult Drosophila brain. Indeed, both neuronal and glial cells responded to MPH and ATX. Further analysis of differentially expressed genes (DEGs) revealed distinct transcriptional changes associated with these two drugs, such as two well-studied dopamine receptor genes (Dop2R and DopEcR) were responsive to MPH but not to ATX at their optimal doses, in addition to genes involved in dopamine metabolism pathways such as Syt1, Sytalpha, Syt7, and Ih in different cell types. More importantly, MPH also suppressed the expression of genes encoding other neurotransmitter receptors and synaptic signaling molecules in many cell types, especially those for Glu and GABA, while the responsive effects of ATX were much weaker. In addition to monoaminergic neuronal transmitters, other neurotransmitters have also shown a similar pattern with respect to a stronger effect associated with MPH than with ATX. Moreover, we identified four distinct glial cell subtypes responsive to the two drugs and detected a greater number of differentially expressed genes associated with ensheathing and astrocyte-like glia. Furthermore, our study provides a rich resource of candidate target genes, supported by drug set enrichment analysis (P = 2.10E-4; hypergeometric test), for the further exploration of drug repurposing. The whole list of candidates can be found at ADHDrug ( http://adhdrug.cibr.ac.cn/ ). In conclusion, we propose a fast and cost-efficient pipeline to explore the underlying molecular mechanisms of ADHD drug treatment in Drosophila brain at single-cell resolution, which may further facilitate drug repurposing applications.

Bioactive Peptide Recognition Based on NLP Pre-Train Algorithm.

Bioactive peptides are defined as peptide sequences within a protein that can regulate important bodily functions through their myriad activities. With the development of machine learning, more computational methods were proposed for bioactive peptides recognition so that this task does not only rely on tedious and time-consuming wet-experiment. But the training and testing process of existing models are limited to small datasets, which affects model performance. Inspired by the success of sequence classification in natural language processing with unlabeled data, we proposed a pre-training method for Bioactive peptides recognition. By pre-trained with large-scale of protein sequences, our method achieved the best performance in multiple functional peptides identification including anti-cancer, anti-diabetic, anti-hypertensive, anti-inflammatory and anti-microbial peptides. Compared with the advanced model, our model's precision, coverage, accuracy and absolute true are improved by 7.2%, 6.9%, 6.1% and 4.2% in the result of 5-fold cross-validation. In addition, the results indicate the model has superior prediction performance in single functional peptides recognition, especially for anti-cancer peptides and anti-microbial peptides which with longer sequences.

DeepTTA: a transformer-based model for predicting cancer drug response.

Identifying new lead molecules to treat cancer requires more than a decade of dedicated effort. Before selected drug candidates are used in the clinic, their anti-cancer activity is generally validated by in vitro cellular experiments. Therefore, accurate prediction of cancer drug response is a critical and challenging task for anti-cancer drugs design and precision medicine. With the development of pharmacogenomics, the combination of efficient drug feature extraction methods and omics data has made it possible to use computational models to assist in drug response prediction. In this study, we propose DeepTTA, a novel end-to-end deep learning model that utilizes transformer for drug representation learning and a multilayer neural network for transcriptomic data prediction of the anti-cancer drug responses. Specifically, DeepTTA uses transcriptomic gene expression data and chemical substructures of drugs for drug response prediction. Compared to existing methods, DeepTTA achieved higher performance in terms of root mean square error, Pearson correlation coefficient and Spearman's rank correlation coefficient on multiple test sets. Moreover, we discovered that anti-cancer drugs bortezomib and dactinomycin provide a potential therapeutic option with multiple clinical indications. With its excellent performance, DeepTTA is expected to be an effective method in cancer drug design.

preMLI: a pre-trained method to uncover microRNA-lncRNA potential interactions.

The interaction between microribonucleic acid and long non-coding ribonucleic acid plays a very important role in biological processes, and the prediction of the one is of great significance to the study of its mechanism of action. Due to the limitations of traditional biological experiment methods, more and more computational methods are applied to this field. However, the existing methods often have problems, such as inadequate acquisition of potential features of the sequence due to simple coding and the need to manually extract features as input. We propose a deep learning model, preMLI, based on rna2vec pre-training and deep feature mining mechanism. We use rna2vec to train the ribonucleic acid (RNA) dataset and to obtain the RNA word vector representation and then mine the RNA sequence features separately and finally concatenate the two feature vectors as the input of the prediction task. The preMLI performs better than existing methods on benchmark datasets and has cross-species prediction capabilities. Experiments show that both pre-training and deep feature mining mechanisms have a positive impact on the prediction performance of the model. To be more specific, pre-training can provide more accurate word vector representations. The deep feature mining mechanism also improves the prediction performance of the model. Meanwhile, The preMLI only needs RNA sequence as the input of the model and has better cross-species prediction performance than the most advanced prediction models, which have reference value for related research.

Population genetic analysis of aquaculture salmonid populations in China using a 57K rainbow trout SNP array.

Various salmonid species are cultivated in cold water aquaculture. However, due to limited genomic data resources, specific high-throughput genotyping tools are not available to many of the salmonid species. In this study, a 57K single nucleotide polymorphism (SNP) array for rainbow trout (Oncorhynchus mykiss) was utilized to detect polymorphisms in seven salmonid species, including Hucho taimen, Oncorhynchus masou, Salvelinus fontinalis, Brachymystax lenok, Salvelinus leucomaenis, O. kisutch, and O. mykiss. The number of polymorphic markers per population ranged from 3,844 (O. kisutch) to 53,734 (O. mykiss), indicating that the rainbow trout SNP array was applicable as a universal genotyping tool for other salmonid species. Among the six other salmonid populations from four genera, 28,882 SNPs were shared, whereas 525 SNPs were polymorphic in all four genera. The genetic diversity and population relationships of the seven salmonid species were studied by principal component analysis (PCA). The phylogenetic relationships among populations were analyzed using the maximum likelihood method, which indicated that the shared SNP markers provide reliable genomic information for population genetic analyses in common aquaculture salmonid fishes. Furthermore, this obtained genomic information may be applicable for population genetic evaluation, marker-assisted breeding, and propagative parent selection in fry production.

Genomic Basis of Adaptive Evolution: The Survival of Amur Ide (Leuciscus waleckii) in an Extremely Alkaline Environment.

The Amur ide (Leuciscus waleckii) is a cyprinid fish that is widely distributed in Northeast Asia. The Lake Dali Nur population inhabits one of the most extreme aquatic environments on Earth, with an alkalinity up to 50 mmol/L (pH 9.6), thus providing an exceptional model with which to characterize the mechanisms of genomic evolution underlying adaptation to extreme environments. Here, we developed the reference genome assembly for L. waleckii from Lake Dali Nur. Intriguingly, we identified unusual expanded long terminal repeats (LTRs) with higher nucleotide substitution rates than in many other teleosts, suggesting their more recent insertion into the L. waleckii genome. We also identified expansions in genes encoding egg coat proteins and natriuretic peptide receptors, possibly underlying the adaptation to extreme environmental stress. We further sequenced the genomes of 10 additional individuals from freshwater and 18 from Lake Dali Nur populations, and we detected a total of 7.6 million SNPs from both populations. In a genome scan and comparison of these two populations, we identified a set of genomic regions under selective sweeps that harbor genes involved in ion homoeostasis, acid-base regulation, unfolded protein response, reactive oxygen species elimination, and urea excretion. Our findings provide comprehensive insight into the genomic mechanisms of teleost fish that underlie their adaptation to extreme alkaline environments.

An ultra-high density linkage map and QTL mapping for sex and growth-related traits of common carp (Cyprinus carpio).

High density genetic linkage maps are essential for QTL fine mapping, comparative genomics and high quality genome sequence assembly. In this study, we constructed a high-density and high-resolution genetic linkage map with 28,194 SNP markers on 14,146 distinct loci for common carp based on high-throughput genotyping with the carp 250 K single nucleotide polymorphism (SNP) array in a mapping family. The genetic length of the consensus map was 10,595.94 cM with an average locus interval of 0.75 cM and an average marker interval of 0.38 cM. Comparative genomic analysis revealed high level of conserved syntenies between common carp and the closely related model species zebrafish and medaka. The genome scaffolds were anchored to the high-density linkage map, spanning 1,357 Mb of common carp reference genome. QTL mapping and association analysis identified 22 QTLs for growth-related traits and 7 QTLs for sex dimorphism. Candidate genes underlying growth-related traits were identified, including important regulators such as KISS2, IGF1, SMTLB, NPFFR1 and CPE. Candidate genes associated with sex dimorphism were also identified including 3KSR and DMRT2b. The high-density and high-resolution genetic linkage map provides an important tool for QTL fine mapping and positional cloning of economically important traits, and improving common carp genome assembly.

Genome wide identification of scavenger receptors class A in common carp (Cyprinus carpio) and their expression following Aeromonas hydrophila infection.

Scavenger receptors class A (SCARAs) is a subgroup of diverse families of pattern recognition receptors that bind a range of ligands, and play important roles in innate immune processes through pathogens detection, adhesion, endocytosis, and phagocytosis. However, most studies of SCARAs have focused on mammals, and much less is known of SCARAs in fish species. In this study, we identified 7 SCARAs across the common carp genome, which were classified into four subclasses according to comparative genomic analysis including sequence similarities analysis, gene structure and functional domain prediction. Further phylogenetic and syntenic analysis supported their annotation and orthologies. Through examining gene copy number of SCARA genes across several vertebrates, SCARA2, SCARA3 and SCARA4 were found have undergone gene duplication. The expression patterns of SCARAs in common carp were examined during early developmental stages, in healthy tissues, and after Aeromonas hydrophila infection. Most SCARA genes were ubiquitously expressed during common carp early developmental stages, and presented diverse patterns in various healthy tissues, with relatively high expression levels in spleen, liver, intestine, gill and brain, indicating their critical roles likely in maintaining homeostasis and host immune response activities. After A. hydrophila infection, most SCARA genes were up-regulated at 4 h post infection in mucosal tissue intestine, while generally up-regulated at 12 h post infection in spleen, suggesting a tissue-specific pattern of regulation. Taken together, all these results suggested that SCARA genes played important roles in host immune response to A. hydrophila infection in common carp, and provided important genomic resources for future studies on fish disease management.

Genome-wide identification, phylogeny, and expression of fibroblast growth genes in common carp.

Fibroblast growth factors (FGFs) are a large family of polypeptide growth factors, which are found in organisms ranging from nematodes to humans. In vertebrates, a number of FGFs have been shown to play important roles in developing embryos and adult organisms. Among the vertebrate species, FGFs are highly conserved in both gene structure and amino-acid sequence. However, studies on teleost FGFs are mainly limited to model species, hence we investigated FGFs in the common carp genome. We identified 35 FGFs in the common carp genome. Phylogenetic analysis revealed that most of the FGFs are highly conserved, though recent gene duplication and gene losses do exist. By examining the copy number of FGFs in several vertebrate genomes, we found that eight FGFs in common carp have undergone gene duplications, including FGF6a, FGF6b, FGF7, FGF8b, FGF10a, FGF11b, FGF13a, and FGF18b. The expression patterns of all FGFs were examined in various tissues, including the blood, brain, gill, heart, intestine, muscle, skin, spleen and kidney, showing that most of the FGFs were ubiquitously expressed, indicating their critical role in common carp. To some extent, examination of gene families with detailed phylogenetic or orthology analysis verified the authenticity and accuracy of assembly and annotation of the recently published common carp whole genome sequences. Gene families are also considered as a unique source for evolutionary studies. Moreover, the whole set of common carp FGF gene family provides an important genomic resource for future biochemical, physiological, and phylogenetic studies on FGFs in teleosts.

Complete mitochondrial genome of yellowfin goby (Acanthogobius hasta).

Yellowfin goby (Acanthogobius hasta) belongs to the Perciformes Gobiidae. In this study, we sequenced the complete mitochondrial genome of A. hasta. The complete mitochondrial genome was determined to be 16,663 bp in length including 13 protein-coding genes, 22 transfer RNAs (tRNAs), 2 ribosomal RNAs (rRNAs) and 1 control region. The complete mitochondrial genome of A. gueldenstaedti provides basic genome data for relative studies on Acipenseriformes.

Complete mitochondrial genomes of two ornamental fishes.

The complete mitochondrial genomes of two ornamental fishes, black molly (Poecilia sphenops) and blue gourami (Trichogaster trichopterus), were obtained by the traditional polymerase chain reaction (PCR)-based sequencing approach. The mitogenomes of P. sphenops and T. trichopterus are determined as 16,533 bp and 16,456 bp in length, respectively. Both the genomes include 22 transfer RNA genes, 13 protein-coding genes and 2 ribosomal RNA genes. Phylogenetic tree was constructed based on the complete mitogenomes of these two species and closely related 20 teleost species to assess their phylogenic relationship and evolution.

Intraspecific mitochondrial variations between Rhinogobio typus from the Yellow River and Yangtze River.

The complete mitochondrial genome of the first individual Rhinogobio typus collected from the Yellow River were sequenced and compared with the previously reported complete mitochondrial sequence of Rhinogobio typus from the Yangtze River. The length of their circular mitochondrial genome was determined to be 16 599 and 16 608 bp respectively. The comparison of two mitochondrial genomes revealed 237 base pair substitutions and 17 insertions or deletions (indels), including 182 base pair substitutions and 2 indels in protein-coding region. Phylogenetic tree was constructed based on complete mitogenomes of the two populations and closely related 13 teleost species to assess their phylogenic relationship and evolution.

Construction of the BAC Library of Small Abalone (Haliotis diversicolor) for Gene Screening and Genome Characterization.

The small abalone (Haliotis diversicolor) is one of the most important aquaculture species in East Asia. To facilitate gene cloning and characterization, genome analysis, and genetic breeding of it, we constructed a large-insert bacterial artificial chromosome (BAC) library, which is an important genetic tool for advanced genetics and genomics research. The small abalone BAC library includes 92,610 clones with an average insert size of 120 Kb, equivalent to approximately 7.6× of the small abalone genome. We set up three-dimensional pools and super pools of 18,432 BAC clones for target gene screening using PCR method. To assess the approach, we screened 12 target genes in these 18,432 BAC clones and identified 16 positive BAC clones. Eight positive BAC clones were then sequenced and assembled with the next generation sequencing platform. The assembled contigs representing these 8 BAC clones spanned 928 Kb of the small abalone genome, providing the first batch of genome sequences for genome evaluation and characterization. The average GC content of small abalone genome was estimated as 40.33%. A total of 21 protein-coding genes, including 7 target genes, were annotated into the 8 BACs, which proved the feasibility of PCR screening approach with three-dimensional pools in small abalone BAC library. One hundred fifty microsatellite loci were also identified from the sequences for marker development in the future. The BAC library and clone pools provided valuable resources and tools for genetic breeding and conservation of H. diversicolor.

Phylogenetic and Evolutionary Analyses of the Frizzled Gene Family in Common Carp (Cyprinus carpio) Provide Insights into Gene Expansion from Whole-Genome Duplications.

In humans, the frizzled (FZD) gene family encodes 10 homologous proteins that commonly localize to the plasma membrane. Besides being associated with three main signaling pathways for cell development, most FZDs have different physiological effects and are major determinants in the development process of vertebrates and. Here, we identified and annotated the FZD genes in the whole-genome of common carp (Cyprinus carpio), a teleost fish, and determined their phylogenetic relationships to FZDs in other vertebrates. Our analyses revealed extensive gene duplications in the common carp that have led to the 26 FZD genes that we detected in the common carp genome. All 26 FZD genes were assigned orthology to the 10 FZD genes of on-land vertebrates, with none of genes being specific to the fish lineage. We postulated that the expansion of the FZD gene family in common carp was the result of an additional whole genome duplication event and that the FZD gene family in other teleosts has been lost in their evolution history with the reason that the functions of genes are redundant and conservation. Through the expression profiling of FZD genes in common carp, we speculate that the ancestral gene was likely capable of performing all functions and was expressed broadly, while some descendant duplicate genes only performed partial functions and were specifically expressed at certain stages of development.

A rubella outbreak investigation and BRD-II strain rubella vaccine effectiveness study, Harbin city, Heilongjiang province, China, 2010-2011.

BACKGROUND: In the 1990s, China introduced rubella vaccine into the private market using BRD-II virus strain, which is different than the globally used RA27/3 strain. In 2007, BRD-II rubella containing vaccine was introduced into the national immunization program and recommended for routine use. However, to our knowledge, there are no field vaccine effectiveness (VE) studies of BRD-II rubella vaccine. In April 2011, a rubella outbreak was detected in two daycare centers in Harbin city, China. We conducted an investigation to determine VE of BRD-II rubella vaccine. METHODS: Rubella cases were either laboratory-confirmed or epidemiologically linked to laboratory-confirmed cases. We collected demographic characteristics, migrant status, and history of rubella and measles vaccination from all children in the two daycare centers. RESULTS: The first case of rubella was on 22 November, 2010. Among the 143 children in the two daycare centers, 22 acquired rubella, for an overall attack rate (AR) of 15.4% (22/143). The AR in higher-grade classes (21.7%) was higher than in lower grade classes (3.9%). The AR among migrant children (47.8%) was higher than among local children (9.2%). Rubella vaccine coverage was 17% (24/143), while measles vaccine coverage was 100%. The AR among rubella-vaccinated children was 0% (0/24), and the AR among rubella-unvaccinated children was 18.5% (22/119), for a VE of 100% (P value=0.025, 95% CI: 35-100%). Rubella vaccine coverage among children born before 2007 was 10.2% (10/98), and was lower than that for children born in 2007 or after (31.1% (14/45), RR=0.33, 95%CI: 0.16-0.68). Emergency vaccination was conducted on 11 and 12 April 2011, and the outbreak stopped in one week later. CONCLUSIONS: Domestic BRD-II strain rubella vaccine showed high vaccine effectiveness against rubella. Rubella vaccine coverage through routine immunization was insufficient. Consideration should be given for measuring rubella vaccine coverage to determine the need for catch-up vaccination in China.