Biology subjects

Luo, X.

Publications and source records attributed to Luo, X..

At least 199 records · Page 11Linked to original sources

bioRxiv2020

Efficient and stable metabarcoding sequencing from DNBSEQ-G400 sequencer examined by large fungal community analysis

ABSTRACTMetabarcoding has become the de facto method for characterizing the structure of microbial communities in complex environmental samples. To determine how sequencing platform may influence microbial community characterization, we present a large-scale comparison of two sequencing platforms; Illumina MiSeq and a new platform DNBSEQ-G400 developed by MGI Tech. The accuracy of DNBSEQ-G400 on bacterial and fungal mock samples and compared sequencing consistency and precision between DNBSEQ-G400 and MiSeq platforms by sequencing the fungal ITS2 region from 1144 soil samples with 3 technical replicates. The DNBSEQ-G400 showed a high accuracy in reproducing mock communities containing different proportions of bacteria and fungi, respectively. The taxonomic profiles of the 1144 soil samples generated by the two DNBSEQ-G400 modes closely resembled each other and were highly correlated with those generated by the MiSeq platform. Analyses of technical replicates demonstrated a run bias against certain taxa on the MiSeq but not DNBSEQ-G400 platform. Based on lower cost, greater capacity, and less bias, we conclude that DNBSEQ-G400 is an optimal platform for short-term metabarcoding of microbial communities.IMPORTANCE Experimental steps that generate sequencing bias during amplicon sequencing have been intensively evaluated, including the choice of primer pair, polymerase, PCR cycle and technical replication. However, few studies have assessed the accuracy and precision of different sequencing platforms. Here, we compared the performance of newly released DNBSEQ-G400 sequencer with that of the commonly used Illumina MiSeq platform by leveraging amplicon sequencing of a large number of soil samples. Significant sequencing bias among major fungal genera was found in parallel MiSeq runs, which can be easily neglected without the use of sequencing controls. We emphasize the importance of technical controls in large-scale sequencing efforts and provide DNBSEQ-G400 as an alternative with increased sequencing capacity and more stable reproducibility for amplicon sequencing.View Full Text

microbiology↗

bioRxiv2020

The Atlas of the In Vivo HIV CD4 T Cell Reservoir

ABSTRACTThe latent reservoir is a main barrier for curing HIV. But because latently-infected cells cannot be phenotyped directly, the features of the in vivo reservoir have remained elusive. Here, we describe a method that leverages high-dimensional phenotyping using CyTOF to trace latently-infected cells reactivated ex vivo to their original pre-activation states. Our results suggest that contrary to common assumptions, the reservoir is not randomly distributed among cell subsets, and is remarkably conserved between individuals. However, reservoir composition differs between tissues and blood, as do cells successfully reactivated by different latency reversing agents. Most importantly, by selecting 8-10 of our 39 original CyTOF markers, we were able to isolate highly purified populations of unstimulated in vivo latent cells, thereby validating the PP-SLIDE approach for reservoir characterization. These purified populations were highly enriched for replication-competent and intact provirus, transcribed HIV, and displayed clonal expansion. The ability to isolate unstimulated latent cells from infected individuals enables previously impossible studies of HIV persistence.Competing Interest StatementThe authors have declared no competing interest.View Full Text

immunology↗

bioRxiv2020

SARS-CoV-2-specific T cells exhibit unique features characterized by robust helper function, lack of terminal differentiation, and high proliferative potential

Convalescing COVID-19 patients mount robust T cell responses against SARS-CoV-2, suggesting an important role for T cells in viral clearance. To date, the phenotypes of SARS-CoV-2-specific T cells remain poorly defined. Using 38-parameter CyTOF, we phenotyped longitudinal specimens of SARS-CoV-2-specific CD4+ and CD8+ T cells from nine individuals who recovered from mild COVID-19. SARS-CoV-2-specific CD4+ T cells were exclusively Th1 cells, and predominantly Tcm with phenotypic features of robust helper function. SARS-CoV-2-specific CD8+ T cells were predominantly Temra cells in a state of less terminal differentiation than most Temra cells. Subsets of SARS-CoV-2-specific T cells express CD127, can homeostatically proliferate, and can persist for over two months. Our results suggest that long-lived and robust T cell immunity is generated following natural SARS-CoV-2 infection, and support an important role for SARS-CoV-2-specific T cells in host control of COVID-19.

immunology↗

bioRxiv2020

Neural Stimulation in vitro and in vivo by Photoacoustic Nanotransducers

Neuromodulation is an invaluable approach for study of neural circuits and clinical treatment of neurological diseases. Here, we report semiconducting polymer nanoparticles based photoacoustic nanotransducers (PANs) for neural stimulation. Our PANs strongly absorb light in the near-infrared second window and generate localized acoustic waves. PANs can also be surface-modified to selectively bind onto neurons. PAN-mediated activation of primary neurons in vitro is achieved with ten 3-nanosecond laser pulses at 1030 nm over a 3 millisecond duration. In vivo neural modulation of mouse brain activities and motor activities is demonstrated by PANs directly injected into brain cortex. With millisecond-scale temporal resolution, sub-millimeter spatial resolution and negligible heat deposition, PAN stimulation is a new non-genetic method for precise control of neuronal activities, opening potentials in non-invasive brain modulation.

neuroscience↗

bioRxiv2020

Superior Colliculus-Projected GABAergic Retinal Ganglion Cells Mediate Looming-Evoked Flight Response

The looming stimulus-evoked flight response is an experimental paradigm for studying innate defensive behaviors. However, how the visual looming stimulus is transmitted from the retina to the brain remains poorly understood. Here, we report that superior colliculus (SC)-projecting RGCs transmit the looming signal from the retina to the brain to mediate the looming-evoked flight behavior by releasing GABA. In the mouse retina, GABAergic RGCs are capable of projecting to many brain areas, including the SC. Superior colliculus (SC)-projecting GABAergic RGCs (spgRGCs) are mono-synaptically connected to the parvalbumin-positive SC neurons known to be required for the looming-evoked flight response. Optogenetic activation of spgRGCs triggers GABA-mediated inhibition in SC neurons. The ablation or silence of spgRGCs compromises looming-evoked flight response but not image-forming functions. Therefore, this study shows that spgRGCs control the looming-evoked flight response by regulating SC neurons via GABA, providing novel insight into the regulation of innate defensive behaviors.

neuroscience↗

bioRxiv2020

Labeling subcellular structures in living specimens using live-cell incompatible dyes with excellent optical properties

Despite the urgent needs of imaging living specimens for cutting-edge biological research, most of the existing fluorescent labeling methods suffer from either poor optical properties or complicated operations to realize cell-permeability and specificity. Here, we introduce a method to overcome this tradeoff by incubating living cells and tissues with fluorescent dyes, no matter if they are cell-permeable or not, at particular conditions (concentration and temperature) without physical cell-penetration or chemical modifications. Based on this method, the mitochondrial labeling capability of Atto dyes, especially Atto 647N with extraordinary optical properties, together with interesting interactions between organelles is revealed. These results indicate the great potential of using dyes, which are normally considered "live-cell incompatible", to capture the morphology and dynamics of subcellular structures in living specimens.

cell biology↗

bioRxiv2020

SIRT1 ameliorates premature senescence-induced defenestration in hepatic sinusoidal endothelial cell

Premature senescence, linked to progerin, involves in endothelial dysfunction and liver diseases. Activating sirtuin 1 (SIRT1) ameliorates liver fibrosis. However, the potential mechanisms of premature senescence in defenestration in hepatic sinusoidal endothelial cells (HSECs) and how SIRT1 affects fenestrae remains elusive. Our study showed that in vivo, premature senescence occurred, with decrease of SIRT1, during CCl4-induced defenestration in HSECs and liver fibrogenesis; whereas overexpressing SIRT1 with adenovirus vector lessened progerin-associated premature senescence to relieve CCl4-induced defenestration and liver fibrosis. In vitro, fenestrae in HSECs disappeared, with progerin-associated premature senescence; these effects aggravated by H2O2-induced oxidative damage. Nevertheless, knockdown of NOX2 or overexpression of SIRT1 with adenovirus vector reduced progerin-associated premature senescence to maintain fenestrae through deacetylating p53. Furthermore, more Ac p53 K381 and progerin co-localized with accumulation of actin filament (F-actin) in the nuclear envelope of H2O2-treated HSECs; in contrast, these effects were rescued by overexpressing SIRT1. In conclusion, NOX2-dependent oxidative damage aggravates defenestration in HSECs via progerin-associated premature senescence; SIRT1-mediated deacetylation of p53 maintains fenestrae and attenuates liver fibrogenesis through inhibiting premature senescence.

cell biology↗

bioRxiv2020

Integrating Genomic Correlation Structure Improves Copy Number Variations Detection

Copy number variation plays important roles in human complex diseases. The detection of copy number variants (CNVs) is identifying mean shift in genetic intensities to locate chromosomal breakpoints, the step of which is referred to as chromosomal segmentation. Many segmentation algorithms have been developed with a strong assumption of independent observations in the genetic loci, and they assume each locus has an equal chance to be a breakpoint (i.e., boundary of CNVs). However, this assumption is violated in the genetics perspective due to the existence of correlation among genomic positions such as linkage disequilibrium (LD). Our study showed that the LD structure is related to the location distribution of CNVs which indeed presents a non-random pattern on the genome. To generate more accurate CNVs, we therefore proposed a novel algorithm, LDcnv, that models the CNV data with its biological characteristics relating to genetic correlation (i.e., LD). To evaluate the performance of LDcnv, we conducted extensive simulations and analyzed large-scale HapMap datasets. We showed that LDcnv presents high accuracy, stability and robustness in CNV detection and higher precision in detecting short CNVs compared to existing methods. We also theoretically demonstrated the correlation structure of CNV data, which further supports the necessity of integrating biological structure in statistical methods for CNV detection. This new segmentation algorithm has a wide scope of application with next-generation sequencing data analysis and single-cell sequencing analysis. Author SummaryCopy number variants (CNVs) refers to gains or losses of the DNA segments in comparison to a reference genome. CNVs have garnered extensive interests in recent years as they play an important role susceptibility to disorders and diseases such as autism, schizophrenia and cancer [1-7]. Although innovation in modern technology is promoting the discoveries related to CNVs, the methodology for CNV detection is still lagging, which limits the novel discoveries regarding the role of CNVs in complex diseases. In this study, we are proposing a novel segmentation algorithm, LDcnv, to accurately locate the breakpoints or boundaries of CNVs in the human genome. Instead of utilizing an independent assumption of the signal intensities as has been used in traditional segmentation algorithms, LDcnv models the correlation structure in the genome in a change-point CNV detection model, which allows for accurate and fast computation with a whole genome scan. Our study showed strong theoretical evidence of the existence of correlation structure in real CNV data, and we believe that taking this evidence into consideration will improve the power of CNV detection. Extensive simulation studies have demonstrated the advantage of the LDcnv algorithm in stability, robustness and accuracy over existing methods. We also used high-quality CNV profiles to further support the superior performance of the LDcnv algorithm over existing methods. The development of the LDcnv algorithm provides great insights for new directions in developing CNV detection tools.

genetics↗

bioRxiv2020

Computational target fishing by mining transcriptional data using a novel Siamese spectral-based graph convolutional network

Computational target fishing aims to investigate the mechanism of action or the side effects of bioactive small molecules. Unfortunately, conventional ligand-based computational methods only explore a confined chemical space, and structure-based methods are limited by the availability of crystal structures. Moreover, these methods cannot describe cellular context-dependent effects and are thus not useful for exploring the targets of drugs in specific cells. To address these challenges, we propose a novel Siamese spectral-based graph convolutional network (SSGCN) model for inferring the protein targets of chemical compounds from gene transcriptional profiles. Although the gene signature of a compound perturbation only provides indirect clues of the interacting targets, the SSGCN model was successfully trained to learn from known compound-target pairs by uncovering the hidden correlations between compound perturbation profiles and gene knockdown profiles. Using a benchmark set, the model achieved impressive target inference results compared with previous methods such as Connectivity Map and ProTINA. More importantly, the powerful generalization ability of the model observed with the external LINCS phase II dataset suggests that the model is an efficient target fishing or repositioning tool for bioactive compounds.

bioinformatics↗

bioRxiv2020

Facing small and biased data dilemma in drug discovery with federated learning

Artificial intelligence (AI) models usually require large amounts of high-quality training data, which is in striking contrast to the situation of small and biased data faced by current drug discovery pipelines. The concept of federated learning has been proposed to utilize distributed data from different sources without leaking sensitive information of these data. This emerging decentralized machine learning paradigm is expected to dramatically improve the success of AI-powered drug discovery. We here simulate the federated learning process with 7 aqueous solubility datasets from different sources, among which there are overlapping molecules with high or low biases in the recorded values. Beyond the benefit of gaining more data, we also demonstrate federated training has a regularization effect making it superior than centralized training on the pooled datasets with high biases. Further, two more cases are studied to test the usability of federated learning in drug discovery. Our work demonstrates the application of federated learning in predicting drug related properties, but also highlights its promising role in addressing the small data and biased data dilemma in drug discovery.

pharmacology and toxicology↗

bioRxiv2020

The evolving systemic biomarker milieu in obese ZSF1 rat model of human cardiometabolic syndrome: Characterization of the model and cardioprotective effect of GDF15

Cardiometabolic syndrome has become a global health issue. Heart failure is a common comorbidity of cardiometabolic syndrome. Successful drug development to prevent cardiometabolic syndrome and associated comorbidities requires preclinical models predictive of human conditions. To characterize the heart failure component of cardiometabolic syndrome, cardiometabolic, metabolic, and renal biomarkers were evaluated in obese and lean ZSF1 20-to 22-week-old male rats. Cardiac function, exercise capacity, and left ventricular gene expression were also analyzed. Obese ZSF1 rats exhibited multiple features of human cardiometabolic syndrome by pathological changes in systemic renal, metabolic, and cardiovascular disease circulating biomarkers. Hemodynamic assessment, echocardiography, and decreased exercise capacity confirmed heart failure with preserved ejection fraction. RNA-seq results demonstrated changes in left ventricular gene expression associated with fatty acid and branched chain amino acid metabolism, cardiomyopathy, cardiac hypertrophy, and heart failure. Twelve weeks of growth differentiation factor 15 (GDF15) treatment significantly decreased body weight, food intake, blood glucose, and triglycerides and improved exercise capacity in obese ZSF1 males. Systemic cardiovascular injury markers were significantly lower in GDF15-treated obese ZSF1 rats. Obese ZSF1 male rats represent a preclinical model for human cardiometabolic syndrome with established heart failure with preserved ejection fraction. GDF15 treatment mediated dietary response and demonstrated a cardioprotective effect in obese ZSF1 rats.

physiology↗

bioRxiv2020

Multiple-biological matrices metabolomics identified new metabolite biomarkers for the precise diagnosis of pancreatic cancer and associated tissue metastasis

PurposeTo improve clinical diagnosis and enhance therapeutic outcome, we figure out to identify and validate metabolite biomarkers from the plasma samples of patients with pancreatic cancer that can easily, sensitively and efficiently diagnose the onsite progression, and metastasis of the disease. Experimental DesignWe employed the newly developed precision-targeted metabolomics method to validate that many differential metabolites have the capacity to markedly distinguish patients with pancreatic cancer from healthy controls. To further enhance the specificity and selectivity of metabolite biomarkers, a dozen tumor tissues from PC patients and paired normal tissues were used to clinically validate the biomarker performance. ResultsWe eventually verified five new metabolite biomarkers in plasma (creatine, inosine, beta-sitosterol, sphinganine and glycocholic acid), which can be used to readily diagnose pancreatic cancer in a clinical setting. Excitingly, we proposed a panel biomarker by integrating these five individual metabolites into one pattern, demonstrating much higher accuracy and specificity to precisely diagnose pancreatic cancer than conventional biomarkers (CA125, CA19-9, CA242 and CEA); Moreover, we characterized succinic acid and gluconic acid as having a great capability to monitor the progression and metastasis of pancreatic cancer at different stages. ConclusionsTaken together, this metabolomics method was used to identify and validate metabolite biomarkers that can precisely and sensitively diagnose the onsite progression and metastasis of pancreatic cancer in a clinical setting. Furthermore, such effort should leave clinicians with the correct time frame to facilitate early and efficiently therapeutic interventions, which could largely improve the five-year survival rate of PC patients.

biochemistry↗

bioRxiv2020

Novel Functional metabolites that affect biofilm formation are regulated by bioavailable iron with siderophore-dependent pathway

Biofilms are broadly formed by diverse microorganisms under stressful environments and are basically surrounded by an EPS matrix, enabling bacterial cells to confer more resistance to biocides, antibiotics and other invasions than their planktonic counterparts. However, biofilm formation causes problems in various fields, including clinical infections, environmental pollution, agricultural production and industrial contamination. Unfortunately, the mechanism of biofilm formation has not been completely elucidated, and currently, we lack an efficient strategy to tackle these tough problems and destroy biofilms. In the present study, we sought to decipher the mechanism of biofilm formation through the regulation of functional metabolites regulated by iron. By exposing bacterial cells to various concentrations of iron, we found that iron can regulate biofilm formation, and phenotypic changes were obviously dependent on iron concentration. A functional metabolome assay was further implemented to investigate the regulatory mechanism of iron on biofilm formation; we verified that siderophores (linear enterobactin, yersiniabactin, di-glucosylated-salmochelin and HPTT-COOH) mostly account for the transportation of iron into bacterial cells. Then, bioavailable iron was recruited by bacterial cells to direct the biosynthesis and expression of five functional metabolites (L-tryptophan, 5-MTA, spermidine, CMP and L-leucine), which were identified as new effectors that directly regulate biofilm formation. Taken together, this study is the first to identify five new metabolic effectors to efficiently regulate biofilm formation, the biosynthesis and expression of these functional metabolites can be targeted to tackle the challenging problems associated with biofilm formation in different fields.

microbiology↗

bioRxiv2020

Arid1a loss potentiates pancreatic β-cell regeneration through activation of EGF signaling

The dynamic regulation of {beta}-cell abundance is poorly understood. Since chromatin remodeling plays critical roles in liver regeneration, these mechanisms could be generally important for regeneration in other tissues. Here we show that the ARID1A mammalian SWI/SNF complex subunit is a critical regulator of {beta}-cell regeneration. Arid1a is highly expressed in quiescent {beta}-cells but is physiologically suppressed when {beta}-cells proliferate during pregnancy or after pancreas resection. Whole-body Arid1a knockout mice were protected against streptozotocin induced diabetes. Cell-type and temporally specific genetic dissection showed that {beta}-cell specific Arid1a deletion could potentiate {beta}-cell regeneration in multiple contexts. Transcriptomic and epigenomic profiling of mutant islets revealed increased Neuregulin-ERBB-NR4A signaling. Functionally, ERBB3 overexpression in {beta}-cells was sufficient to protect against diabetes, and chemical inhibition of ERBB or NR4A was able to block increased regeneration associated with Arid1a loss. mSWI/SNF complex activity is a barrier to {beta}-cell regeneration in physiologic and disease states.

genetics↗

bioRxiv2020

Anacardic acid protects against phenylephrine-induced mouse cardiac hypertrophy through JNK signaling-dependent regulation of histone acetylation

Cardiac hypertrophy is a complex process induced by the activation of multiple signaling pathways. We previously reported that anacardic acid (AA), a histone acetylase (HAT) inhibitor, attenuates phenylephrine (PE)-induced cardiac hypertrophy by downregulating histone H3 acetylation at lysine 9 (H3K9ac). Unfortunately, the upstream signaling events remained unknown. The mitogen-activated protein kinase (MAPK) pathway is an important regulator of cardiac hypertrophy. In this study, we explored the role of JNK/MAPK signaling in cardiac hypertrophy. A mouse model of cardiomyocyte hypertrophy was successfully established in vitro using PE. This study showed that p-JNK directly interacts with HATs (P300 and P300/CBP-associated factor, PCAF) and alters H3K9ac. In addition, both the JNK inhibitor SP600125 and the HAT inhibitor AA attenuated p-JNK overexpression and H3K9 hyperacetylation by inhibiting P300 and PCAF during PE-induced cardiomyocyte hypertrophy. Moreover, we demonstrated that both SP600125 and AA attenuate the overexpression of cardiac hypertrophy-related genes (MEF2A, ANP, BNP, and {beta}-MHC), preventing cardiomyocyte hypertrophy and dysfunction. These results revealed a novel mechanism through which AA might protect mice from PE-induced cardiac hypertrophy. In particular, AA inhibits the effects of JNK signaling on HAT-mediated histone acetylation, and could therefore be used to prevent and treat hypertrophic cardiomyopathy.

cell biology↗

bioRxiv2020

A Whole-Brain Regression Method to Identify Individual and Group Variations in Functional Connectivity

Resting-state functional connectivity is an important and widely used measure of individual and group differences. These differences are typically attributed to various demographic and/or clinical factors. Yet, extant statistical methods are limited to linking covariates with variations in functional connectivity across subjects, especially at the voxel-wise level of the whole brain. This paper introduces a generalized linear model method that regresses whole-brain functional connectivity on covariates. Our approach builds on two methodological components. We first employ whole-brain group ICA to reduce the dimensionality of functional connectivity matrices, and then search for matrix variations associated with covariates using covariate assisted principal regression, a recently introduced covariance matrix regression method. We demonstrate the efficacy of this approach using a resting-state fMRI dataset of a medium-sized cohort of subjects obtained from the Human Connectome Project. The results show that the approach enjoys improved statistical power in detecting interaction effects of sex and alcohol on whole-brain functional connectivity, and in identifying the brain areas contributing significantly to the covariate-related differences in functional connectivity.

neuroscience↗

bioRxiv2019

Semiparametric Partial Common Principal Component Analysis for Covariance Matrices

SO_SCPLOWUMMARYC_SCPLOWWe consider the problem of jointly modeling multiple covariance matrices by partial common principal component analysis (PCPCA), which assumes a proportion of eigenvectors to be shared across covariance matrices and the rest to be individual-specific. This paper proposes consistent estimators of the shared eigenvectors in PCPCA as the number of matrices or the number of samples to estimate each matrix goes to infinity. We prove such asymptotic results without making any assumptions on the ranks of eigenvalues that are associated with the shared eigenvectors. When the number of samples goes to infinity, our results do not require the data to be Gaussian distributed. Furthermore, this paper introduces a sequential testing procedure to identify the number of shared eigenvectors in PCPCA. In simulation studies, our method shows higher accuracy in estimating the shared eigenvectors than competing methods. Applied to a motor-task functional magnetic resonance imaging data set, our estimator identifies meaningful brain networks that are consistent with current scientific understandings of motor networks during a motor paradigm.

neuroscience↗

bioRxiv2019

Both Determinants of Allosteric and Active Sites Responsible for Catalytic Activity of Delta 12 Fatty Acid Desaturase by Domain Swapping

Cheese lacks essential fatty acids (EFAs). Delta 12 fatty acid desaturase (FADS12) is a critical enzyme required for EFA biosynthesis in fermentation of the predominant strains of cheese. Previously, we identified the FADS12 gene and characterized its function for the first time in Geotrichum candidum, a dominant strain used to manufacture soft cheese with white rind. In this study, we analyzed the molecular mechanism of FADS12 function by swapping domains from Mortierella alpina and G. candidum that had, respectively, high and low oleic acid conversion rates. The results revealed three regions that are essential to this process, including regions from the end of the second transmembrane domain to the beginning of the third transmembrane domain, from the end of the third transmembrane domain to the beginning of the fourth transmembrane domain, and from the 30-amino acid from the end of the sixth transmembrane domain to the C-terminal end region. Based on our domain swapping analyses, nine pairs of amino acids including H112, S118, H156, Q161, K301, R306, E307, A309 and S323 in MaFADS12 (K123, A129, N167, M172, T302, D307, I308, E310 and D324 in GcFADS12) were identified as having a significantly effect on FADS12 catalytic efficiency, and linoleic acid and its analogues (12,13-cyclopropenoid fatty acid) were found to inhibit the catalytic activity of FADS12 and related recombinant enzymes. Furthermore, the molecular mechanism of FADS12 inhibition was analyzed. The results revealed two allosteric domains, including one domain from the N-terminal region to the beginning of the first transmembrane domain and another from the 31st amino acid from the end of the sixth transmembrane domain to the C terminus. Y4 and F398 amino acid residues from MaFADS12 and eight pairs of amino acids including G56, L60, L344, G10, Q13, S24, K326 and L344 in MaFADS12 (while Y66, F70, F345, F20, Y23, Y34, F327 and F345 in GcFADS12) played a pivotal role in FADS12 inhibition. Finally, we found that both allosteric and active sites were responsible for the catalytic activity of FADS12 at various temperatures, pH, and times. This study offers a solid theoretical basis to develop preconditioning methods to increase the rate at which GcFADS12 converts oleic and linoleic acids to produce higher levels of EFAs in cheese.

molecular biology↗