Repertoires of tRNAs: The Couplers of Genomics and Proteomics.

The pool of transfer RNA (tRNA) molecules in cells allows the ribosome to decode genetic information. This repertoire of molecular decoders is positioned in the crossroad of the genome, the transcriptome, and the proteome. Omics and systems biology now allow scientists to explore the entire repertoire of tRNAs of many organisms, revealing basic exciting biology. The tRNA gene set of hundreds of species is now characterized, in addition to the tRNA genes of organelles and viruses. Genes encoding tRNAs for certain anticodon types appear in dozens of copies in a genome, while others are universally absent from any genome. Transcriptome measurement of tRNAs is challenging, but in recent years new technologies have allowed researchers to determine the dynamic expression patterns of tRNAs. These advances reveal that availability of ready-to-translate tRNA molecules is highly controlled by several transcriptional and posttranscriptional regulatory processes. This regulation shapes the proteome according to the cellular state. The tRNA pool profoundly impacts many aspects of cellular and organismal life, including protein expression level, translation accuracy, adequacy of folding, and even mRNA stability. As a result, the shape of the tRNA pool affects organismal health and may participate in causing conditions such as cancer and neurological conditions.

PaxDb 5.0: Curated Protein Quantification Data Suggests Adaptive Proteome Changes in Yeasts.

The "Protein Abundances Across Organisms" database (PaxDb) is an integrative metaresource dedicated to protein abundance levels, in tissue-specific or whole-organism proteomes. PaxDb focuses on computing best-estimate abundances for proteins in normal/healthy contexts and expresses abundance values for each protein in "parts per million" in relation to all other protein molecules in the cell. The uniform data reprocessing, quality scoring, and integrated orthology relations have made PaxDb one of the preferred tools for comparisons between individual datasets, tissues, or organisms. In describing the latest version 5.0 of PaxDb, we particularly emphasize the data integration from various types of raw data and how we expanded the number of organisms and tissue groups as well as the proteome coverage. The current collection of PaxDb includes 831 original datasets from 170 species, including 22 Archaea, 81 Bacteria, and 67 Eukaryota. Apart from detailing the data update, we also present a comparative analysis of the human proteome subset of PaxDb against the two most widely used human proteome data resources: Human Protein Atlas and Genotype-Tissue Expression. Lastly, through our protein abundance data, we reveal an evolutionary trend in the usage of sulfur-containing amino acids in the proteomes of Fungi.

Investigation of Cellular Response to the HSP90 Inhibition in Human Cells Through Thermal Proteome Profiling.

Heat shock proteins are chaperones, and they are responsible for protein folding in cells. Heat shock protein 90 (HSP90) is one of the most important chaperones in human cells, and its inhibition is promising for cancer therapy. However, despite the development of multiple HSP90 inhibitors, none of them has been approved for disease treatment due to unexpected cellular toxicity and side effects. Hence, a more comprehensive investigation of cellular response to HSP90 inhibitors can aid in a better understanding of the molecular mechanisms of the cytotoxicity and side effects of these inhibitors. The thermal stability shifts of proteins, which represent protein structure and interaction alterations, can provide valuable information complementary to the results obtained from commonly used abundance-based proteomics analysis. Here, we systematically investigated cell response to different HSP90 inhibitors through global quantification of protein thermal stability changes using thermal proteome profiling, together with the measurement of protein abundance changes. Besides the targets and potential off-targets of the drugs, proteins with significant thermal stability changes under the HSP90 inhibition are found to be involved in cell stress responses and the translation process. Moreover, proteins with thermal stability shifts under the inhibition are upstream of those with altered expression. These findings indicate that the HSP90 inhibition perturbs cell transcription and translation processes. The current study provides a different perspective for achieving a better understanding of cellular response to chaperone inhibition.

Integrated Proteomics Analysis of Baseline Protein Expression in Pig Tissues.

The availability of an increasingly large amount of public proteomics data sets presents an opportunity for performing combined analyses to generate comprehensive organism-wide protein expression maps across different organisms and biological conditions. Sus scrofa, a domestic pig, is a model organism relevant for food production and for human biomedical research. Here, we reanalyzed 14 public proteomics data sets from the PRIDE database coming from pig tissues to assess baseline (without any biological perturbation) protein abundance in 14 organs, encompassing a total of 20 healthy tissues from 128 samples. The analysis involved the quantification of protein abundance in 599 mass spectrometry runs. We compared protein expression patterns among different pig organs and examined the distribution of proteins across these organs. Then, we studied how protein abundances were compared across different data sets and studied the tissue specificity of the detected proteins. Of particular interest, we conducted a comparative analysis of protein expression between pig and human tissues, revealing a high degree of correlation in protein expression among orthologs, particularly in brain, kidney, heart, and liver samples. We have integrated the protein expression results into the Expression Atlas resource for easy access and visualization of the protein expression data individually or alongside gene expression data.

Evolution of Transcript Abundance is Influenced by Indels in Protein Low Complexity Regions.

Protein Protein low complexity regions (LCRs) are compositionally biased amino acid sequences, many of which have significant evolutionary impacts on the proteins which contain them. They are mutationally unstable experiencing higher rates of indels and substitutions than higher complexity regions. LCRs also impact the expression of their proteins, likely through multiple effects along the path from gene transcription, through translation, and eventual protein degradation. It has been observed that proteins which contain LCRs are associated with elevated transcript abundance (TAb), despite having lower protein abundance. We have gathered and integrated human data to investigate the co-evolution of TAb and LCRs through ancestral reconstructions and model inference using an approximate Bayesian calculation based method. We observe that on short evolutionary timescales TAb evolution is significantly impacted by changes in LCR length, with insertions driving TAb down. But in contrast, the observed data is best explained by indel rates in LCRs which are unaffected by shifts in TAb. Our work demonstrates a coupling between LCR and TAb evolution, and the utility of incorporating multiple responses into evolutionary analyses.

Expression of prolyl hydroxylase domains, the upstream regulators of HIF, in the brain of the anoxia-tolerant crucian carp during anoxia-reoxygenation.

The hypoxia-inducible factor (HIF) is considered key in the transcriptional response to low oxygen. Yet, the role of HIF in the absence of oxygen (anoxia) and in preparation for reoxygenation remains unclear. Recent studies suggest that mounting a HIF response may be counterproductive for anoxia survival. We here studied one of the champions of anoxia survival, the crucian carp (Carassius carassius), and hypothesized that expression of prolyl hydroxylase domains (PHDs; the upstream regulators of HIF) are upregulated to circumvent an energy-costly activation of HIF in anoxia and to prepare for reoxygenation. We measured whole brain mRNA and protein levels of the three isoforms PHD1, PHD2, and PHD3, coded for by multiple paralogs of the genes egln2, egln1, and egln3, using quantitative PCR and Western blotting in the brain of crucian carps exposed to 5 days normoxia or anoxia, and 5 days anoxia followed by 3 or 24 h of reoxygenation. The mRNA levels of most egln paralogs were increased in anoxia and upon reoxygenation, with egln3 showing the largest increase in mRNA level (up to 17-fold) and highest relative mRNA abundance (up to 75% of expressed egln). The protein level of all PHDs was maintained in anoxia and increased upon reoxygenation. We then explored PHD distribution in different brain regions and found PHD immunoreactivity to be associated with axonal branches and showing region-specific changes during anoxia-reoxygenation. Our results support an overall upregulation of egln under prolonged anoxia and PHDs upon reoxygenation in crucian carp, likely aimed at suppressing HIF responses, although regional differences are apparent in such a complex organ as the brain.NEW & NOTEWORTHY We report a profound upregulation of most egln paralog mRNA levels in anoxia and upon reoxygenation, with egln3ii showing the largest, a 17-fold increase, and highest relative mRNA abundance. The relative abundance of prolyl hydroxylase domain (PHD) proteins was maintained during anoxia and increased at reoxygenation. PHD immunoreactivity was localized to axonal branches with region-specific changes during anoxia-reoxygenation. These dynamic and regional changes in crucian carp, champion of anoxia tolerance, are most likely adaptive and call for further mechanistic studies.

Pikeperch muscle tissues: a comparative study of structure, enzymes, genes, and proteins in wild and farmed fish.

Pikeperch (Sander lucioperca) is a freshwater species and an internationally highly demanded fish in aquaculture. Despite intensive research efforts on this species, fundamental knowledge of skeletal muscle biology and structural characteristics is missing. Therefore, we conducted a comprehensive analysis of skeletal muscle parameters in adult pikeperch from two different origins, wild-caught specimens from a lake and those reared in a recirculating aquaculture system. The analyses comprised the biochemical characteristics (nucleic acid, protein content), enzyme activities (creatine kinase, lactate dehydrogenase, NADP-dependent isocitrate dehydrogenase), muscle-specific gene and protein expression (related to myofibre formation, regeneration and permanent growth, muscle structure), and muscle fibre structure. The findings reveal distinct differences between the skeletal muscle of wild and farmed pikeperch. Specifically, nucleic acid content, enzyme activity, and protein expression varied significantly. The higher enzyme activity observed in wild pikeperch suggests greater metabolically activity in their muscles. Conversely, farmed pikeperch indicated a potential for pronounced muscle growth. As the data on pikeperch skeletal muscle characteristics is sparse, the purpose of our study is to gain fundamental insights into the characteristics of adult pikeperch muscle. The presented data serve as a foundation for further research on percids' muscle biology and have the potential to contribute to advancements and adaptations in aquaculture practices.

Comparative proteomic analysis identifies differentially expressed proteins associated with meiotic arrest in cattle-yak hybrids.

Cattle-yak, the interspecific hybrid between yak and taurine cattle, exhibits male-specific sterility. Massive loss of spermatogenic cells, especially spermatocytes, results in azoospermia in these animals. Currently, the mechanisms underlying meiosis block and defects in spermatocyte development remain elusive. The present study was designed to investigate the differences in the protein composition of spermatocytes isolated from 12-month-old yak and cattle-yak testes. Histological analysis confirmed that spermatocytes were the most advanced germ cells in the testes of yak and cattle-yak at this developmental stage. Comparative proteomic analysis identified a total of 452 differentially abundant proteins (DAPs) in the fluorescence-activated cell sorting (FACS) isolated spermatocytes from cattle-yak and yak. A total of 291 proteins were only present in yak spermatocytes. Gene Ontology analysis revealed that the downregulated DAPs were mostly enriched in the cellular response to DNA damage stimulus and double-strand breaks (DSBs) repair via break-induced replication, while the proteins specific for yak were related to cell division and cycle, spermatogenesis, and negative regulation of the extrinsic apoptotic signaling pathway. Ultimately, these DAPs were related to the critical process for spermatocyte meiotic events, including DSBs, homologous recombination, synapsis, crossover formation, and germ cell apoptosis. The database composed of proteins associated with spermatogenesis, including KPNA2, HTATSF1, TRIP12, STIP1, LZTFL1, LARP7, MTCH2, STK31, ROMO1, CDK5AP2, DNMT1, RBM44, and CHRAC1, is the focus of further research on male hybrid sterility. In total, these results provide insight into the molecular mechanisms underlying failed meiotic processes and male infertility in cattle-yak.

Comparison of Database Searching Programs for the Analysis of Single-Cell Proteomics Data.

Single-cell proteomics is emerging as an important subfield in the proteomics and mass spectrometry communities, with potential to reshape our understanding of cell development, cell differentiation, disease diagnosis, and the development of new therapies. Compared with significant advancements in the "hardware" that is used in single-cell proteomics, there has been little work comparing the effects of using different "software" packages to analyze single-cell proteomics datasets. To this end, seven popular proteomics programs were compared here, applying them to search three single-cell proteomics datasets generated by three different platforms. The results suggest that MSGF+, MSFragger, and Proteome Discoverer are generally more efficient in maximizing protein identifications, that MaxQuant is better suited for the identification of low-abundance proteins, that MSFragger is superior in elucidating peptide modifications, and that Mascot and X!Tandem are better for analyzing long peptides. Furthermore, an experiment with different loading amounts was carried out to investigate changes in identification results and to explore areas in which single-cell proteomics data analysis may be improved in the future. We propose that this comparative study may provide insight for experts and beginners alike operating in the emerging subfield of single-cell proteomics.

Deep Profiling of the Proteome Dynamics of Pseudomonas aeruginosa Reference Strain PAO1 under Different Growth Conditions.

As one of the most common bacterial pathogens causing nosocomial infections, Pseudomonas aeruginosa is highly adaptable to survive under various conditions. Here, we profiled the abundance dynamics of 3489 proteins across different growth stages in the P. aeruginosa reference strain PAO1 using data-independent acquisition-based quantitative proteomics. The proteins differentially expressed during the planktonic growth exhibit several distinct patterns of expression profiles and are relevant to various biological processes, highlighting the continuous adaptation of the PAO1 proteome during the transition from the acceleration phase to the stationary phase. By contrasting the protein expressions in a biofilm to planktonic cells, the known roles of T6SS, phenazine biosynthesis, quorum sensing, and c-di-GMP signaling in the biofilm formation process were confirmed. Additionally, we also discovered several new functional proteins that may play roles in the biofilm formation process. Lastly, we demonstrated the general concordance of protein expressions within operons across various growth states, which permits the study of coexpression protein units, and reversely, the study of regulatory components in the operon structure. Taken together, we present a high-quality and valuable resource on the proteomic dynamics of the P. aeruginosa reference strain PAO1, with the potential of advancing our understanding of the overall physiology of Pseudomonas bacteria.

A systems-biology approach identifies co-expression modules in response to low phosphate supply in maize lines of different breeding history.

Plants cope with low phosphorus availability by adjusting growth and metabolism through transcriptomic and proteomic adaptations. We hypothesize that selected genotypes with distinct phosphorous (P) use efficiency covering the breeding history of European Flint heterotic pool provide a tool to reveal general and genotype-specific molecular responses to P limitation. We reconstructed protein and gene co-expression networks by weighted correlation network analysis and related these to phosphate deficiency-induced traits. In roots, low phosphate supply resulted in a decreasing abundance of proteins in the oxidative pentose phosphate pathway and a negative correlation with root and shoot phosphate content. We observed an increase in abundance and positive correlation with root and shoot phosphate content for proteins in sucrose biosynthesis, lipid metabolism, respiration and RNA processing. Purple acid phosphatases, superoxide dismutase and phenylalanine ammonia lyase were identified as being upregulated under low phosphate in all genotypes. Overall, correlations between protein and mRNA abundance changes were limited, with ribosomal proteins and the ubiquitin protein degradation pathway exclusively responding with protein abundance changes. Carbohydrate, phospho- and sulfo-lipid metabolism showed abundance changes at the protein and mRNA levels. These partially non-overlapping proteomic and transcriptomic adjustments to low phosphate suggest sugar and lipid metabolism as metabolic processes associated with improved P use efficiency specifically in Founder Flint lines. We identified a mitogen-activated protein kinase-kinase as a potential genotype-specific regulator of sucrose metabolism at low phosphate in Founder Flint line EP1. We conclude that, during breedingt of Elite Flint lines, regulation of primary metabolism has changed to result in a distinct low phosphate response in Founder lines.

Low Complexity Regions in Mammalian Proteins are Associated with Low Protein Abundance and High Transcript Abundance.

Low Complexity Regions (LCRs) are present in a surprisingly large number of eukaryotic proteins. These highly repetitive and compositionally biased sequences are often structurally disordered, bind promiscuously, and evolve rapidly. Frequently studied in terms of evolutionary dynamics, little is known about how LCRs affect the expression of the proteins which contain them. It would be expected that rapidly evolving LCRs are unlikely to be tolerated in strongly conserved, highly abundant proteins, leading to lower overall abundance in proteins which contain LCRs. To test this hypothesis and examine the associations of protein abundance and transcript abundance with the presence of LCRs, we have integrated high-throughput data from across mammals. We have found that LCRs are indeed associated with reduced protein abundance, but are also associated with elevated transcript abundance. These associations are qualitatively consistent across 12 human tissues and nine mammalian species. The differential impacts of LCRs on abundance at the protein and transcript level are not explained by differences in either protein degradation rates or the inefficiency of translation for LCR containing proteins. We suggest that rapidly evolving LCRs are a source of selective pressure on the regulatory mechanisms which maintain steady-state protein abundance levels.

Comparative protein analysis of two maize genotypes with contrasting tolerance to low temperature.

BACKGROUND: Low temperature (LT) stress is one of the major environmental stress factors affecting the growth and yield of maize (Zea mays L.). Hence, it is important to unravel the molecular mechanisms behind LT stress tolerance to improve molecular breeding in LT tolerant genotypes. In the present study, two maize genotypes viz. Gurez local from Kashmir Himalaya and tropical grown GM6, were dissected for their LT stress response in terms of accumulation of differentially regulated proteins (DRPs). Leaf proteome analysis at three-leaf stage of maize seedlings subjected to LT stress of 6 °C for a total of 12 h duration was performed using two dimensional gel electrophoresis (2D-PAGE) followed by subsequent identification of the proteins involved. RESULTS: After MALDI-TOF (Matrix-assisted laser desorption/ionization-time of flight) and bioinformatics analysis, 19 proteins were successfully identified in Gurez local, while as 10 proteins were found to get successful identification in GM6. The interesting observations from the present investigation is the identification of three novel proteins viz. threonine dehydratase biosynthetic chloroplastic, thylakoidal processing peptidase 1 chloroplastic, and nodulin-like protein, whose role in abiotic stress tolerance, in general, and LT stress, in particular, has not been reported so far. It is important to highlight here that most of LT responsive proteins including the three novel proteins were identified from Gurez local only, owing to its exceptional LT tolerance. From the protein profiles, obtained in both genotypes immediately after LT stress perception, it was inferred that stress responsive protein accumulation and their expression fashion help the Gurez local in seedling establishment and withstand unfavorable conditions as compared to GM6. This was inferred from the findings of pathway enrichment analysis like regulation of seed growth, timing of floral transition, lipid glycosylation, and aspartate family amino acid catabolic processes, besides other key stress defense mechanisms. However, in GM6, metabolic pathways enriched were found to be involved in more general processes including cell cycle DNA replication and regulation of phenylpropanoid metabolism. Furthermore, majority of the qRT-PCR results of the selected proteins demonstrated positive correlation between protein levels and transcript abundance, thereby strengthening our findings. CONCLUSIONS: In conclusion, our findings reported majority of the identified proteins in Gurez local exhibiting up-regulated pattern under LT stress as compared to GM6. Furthermore, three novel proteins induced by LT stress were found in Gurez local, requiring further functional validation. Therefore, our results offer more insights for elucidating the molecular networks mediating LT stress tolerance in maize.

Estimating Total Quantitative Protein Content in Escherichia coli, Saccharomyces cerevisiae, and HeLa Cells.

The continuous improvement of proteomic techniques, most notably mass spectrometry, has generated quantified proteomes of many organisms with unprecedented depth and accuracy. However, there is still a significant discrepancy in the reported numbers of total protein molecules per specific cell type. In this article, we explore the results of proteomic studies of Escherichia coli, Saccharomyces cerevisiae, and HeLa cells in terms of total protein copy numbers per cell. We observe up to a ten-fold difference between reported values. Investigating possible reasons for this discrepancy, we conclude that neither an unmeasured fraction of the proteome nor biases in the quantification of individual proteins can explain the observed discrepancy. We normalize protein copy numbers in each study using a total protein amount per cell as reported in the literature and create integrated proteome maps of the selected model organisms. Our results indicate that cells contain from one to three million protein molecules per µm3 and that protein copy density decreases with increasing organism complexity.

The Cellular Abundance of Chemoreceptors, Chemosensory Signaling Proteins, Sensor Histidine Kinases, and Solute Binding Proteins of Pseudomonas aeruginosa Provides Insight into Sensory Preferences and Signaling Mechanisms.

Chemosensory pathways and two-component systems are important bacterial signal transduction systems. In the human pathogen Pseudomonas aeruginosa, these systems control many virulence traits. Previous studies showed that inorganic phosphate (Pi) deficiency induces virulence. We report here the abundance of chemosensory and two-component signaling proteins of P. aeruginosa grown in Pi deficient and sufficient media. The cellular abundance of chemoreceptors differed greatly, since a 2400-fold difference between the most and least abundant receptors was observed. For many chemoreceptors, their amount varied with the growth condition. The amount of chemoreceptors did not correlate with the magnitude of chemotaxis to their cognate chemoeffectors. Of the four chemosensory pathways, proteins of the Che chemotaxis pathway were most abundant and showed little variation in different growth conditions. The abundance of chemoreceptors and solute binding proteins indicates a sensing preference for amino acids and polyamines. There was an excess of response regulators over sensor histidine kinases in two-component systems. In contrast, ratios of the response regulators CheY and CheB to the histidine kinase CheA of the Che pathway were all below 1, indicative of different signaling mechanisms. This study will serve as a reference for exploring sensing preferences and signaling mechanisms of other bacteria.

Comparative 4D Label-Free Quantitative Proteomic Analysis of Bombus terrestris Provides Insights into Proteins and Processes Associated with Diapause.

Diapause, an adaptative strategy for survival under harsh conditions, is a dynamic multi-stage process. Bombus terrestris, an important agricultural pollinator, is declining in the wild, but artificial breeding is possible by imitating natural conditions. Mated queen bees enter reproductive diapause in winter and recover in spring, but the regulatory mechanisms remain unclear. Herein, we conducted a comparative 4D label-free proteomic analysis of queen bees during artificial breeding at seven timepoints, including pre-diapause, diapause, and post-diapause stages. Through bioinformatics analysis of proteomic and detection of substance content changes, our results found that, during pre-diapause stages, queen bees had active mitochondria with high levels of oxidative phosphorylation, high body weight, and glycogen and TAG content, all of which support energy consumption during subsequent diapause. During diapause stages, body weight and water content were decreased but glycerol increased, contributing to cold resistance. Dopamine content, immune defense, and protein phosphorylation were elevated, while fat metabolism, protein export, cell communication, signal transduction, and hydrolase activity decreased. Following diapause termination, JH titer, water, fatty acid, and pyruvate levels increased, catabolism, synaptic transmission, and insulin signaling were stimulated, ribosome and cell cycle proteins were upregulated, and cell proliferation was accelerated. Meanwhile, TAG and glycogen content decreased, and ovaries gradually developed. These findings illuminate changes occurring in queen bees at different diapause stages during commercial production.

Intramembrane protease RHBDL4 cleaves oligosaccharyltransferase subunits to target them for ER-associated degradation.

The endoplasmic reticulum (ER)-resident intramembrane rhomboid protease RHBDL4 generates metastable protein fragments and together with the ER-associated degradation (ERAD) machinery provides a clearance mechanism for aberrant and surplus proteins. However, the endogenous substrate spectrum and with that the role of RHBDL4 in physiological ERAD is mainly unknown. Here, we use a substrate trapping approach in combination with quantitative proteomics to identify physiological RHBDL4 substrates. This revealed oligosaccharyltransferase (OST) complex subunits such as the catalytic active subunit STT3A as substrates for the RHBDL4-dependent ERAD pathway. RHBDL4-catalysed cleavage inactivates OST subunits by triggering dislocation into the cytoplasm and subsequent proteasomal degradation. RHBDL4 thereby controls the abundance and activity of OST, suggesting a novel link between the ERAD machinery and glycosylation tuning.

Protein speciation is likely to increase the chance of proteins to be determined in 2-DE/MS.

Multiple spotting due to protein speciation might increase a protein's chance of being captured in a random selection of 2-DE spots. We tested this expectation in new (PXD015649) and previously published 2-DE/MS data of porcine and human tissues. For comparison, we included bottom-up proteomics studies (BU-LC/MS) of corresponding biological materials. Analyses of altogether ten datasets proposed that amino acid modification fosters multispotting in 2-DE. Thus, the number of 2-DE spots containing a particular protein more tightly associated with a peptide diversity measure accounting for amino acid modification than with an alternative one disregarding it. Furthermore, every 11th amino acid was a post-translational modification candidate site in 2-DE/MS proteins, whereas in BU-LC/MS proteins this was merely the case in every 21st amino acid. Alternative splicing might contribute to multispotting, since genes encoding 2-DE/MS proteins were found to have on average about 0.3 more transcript variants than their counterparts from BU-LC/MS studies. Correspondingly, resolution completeness as estimated from the representation of transcript variant-rich genes was higher in 2-DE/MS than BU-LC/MS datasets. These findings suggest that the ability to resolve proteomes down to protein species can lead to enrichment of multispotting proteins in 2-DE/MS. Low sensitivity of stains and MS instruments appears to enhance this effect.

Ensemble learning models that predict surface protein abundance from single-cell multimodal omics data.

Single-cell protein abundance is a fundamental type of information to characterize cell states. Due to high cost and technical barriers, however, direct quantification of proteins is difficult. Single-cell RNA sequencing (scRNA-seq) data, serving as a cost-effective substitute of single-cell proteomics, may not accurately reflect protein expression levels due to measurement error, noise, post-transcriptional and translational regulation, etc. The recently emerging single-cell multimodal omics data, e.g. CITE-seq and REAP-seq, can simultaneously profile RNA and protein abundances in single cells, providing labeled data for predictive modeling in a supervised learning framework. Deep neural network-based transfer learning method has been applied to imputation of surface protein abundances from single-cell transcriptomic data. However, it is unclear if the artificial neural network is the best model, and it is desirable to improve the prediction performance (e.g. accuracy, interpretability) of machine learning models. In this paper, we compared several tree-based ensemble learning methods with neural network models, and found that ensemble learning often performed better than neural network, and Random Forest (RF) performed the best overall. Moreover, we used the feature importance scores from RF to interpret biological mechanisms underlying the prediction. Our study demonstrates the effectiveness of ensemble learning for reliable protein abundances prediction using single-cell multimodal omics data, and paves the way for knowledge discovery by mining single-cell multi-omics data in large scale.

Bioinformatic Assessment of Factors Affecting the Correlation between Protein Abundance and Elongation Efficiency in Prokaryotes.

Protein abundance is crucial for the majority of genetically regulated cell functions to act properly in prokaryotic organisms. Therefore, developing bioinformatic methods for assessing the efficiency of different stages of gene expression is of great importance for predicting the actual protein abundance. One of these steps is the evaluation of translation elongation efficiency based on mRNA sequence features, such as codon usage bias and mRNA secondary structure properties. In this study, we have evaluated correlation coefficients between experimentally measured protein abundance and predicted elongation efficiency characteristics for 26 prokaryotes, including non-model organisms, belonging to diverse taxonomic groups The algorithm for assessing elongation efficiency takes into account not only codon bias, but also number and energy of secondary structures in mRNA if those demonstrate an impact on predicted elongation efficiency of the ribosomal protein genes. The results show that, for a number of organisms, secondary structures are a better predictor of protein abundance than codon usage bias. The bioinformatic analysis has revealed several factors associated with the value of the correlation coefficient. The first factor is the elongation efficiency optimization type-the organisms whose genomes are optimized for codon usage only have significantly higher correlation coefficients. The second factor is taxonomical identity-bacteria that belong to the class Bacilli tend to have higher correlation coefficients among the analyzed set. The third is growth rate, which is shown to be higher for the organisms with higher correlation coefficients between protein abundance and predicted translation elongation efficiency. The obtained results can be useful for further improvement of methods for protein abundance prediction.