scholarly journals Discovery of Novel Conotoxin Candidates Using Machine Learning

Toxins ◽  
2018 ◽  
Vol 10 (12) ◽  
pp. 503 ◽  
Author(s):  
Qing Li ◽  
Maren Watkins ◽  
Samuel Robinson ◽  
Helena Safavi-Hemami ◽  
Mark Yandell
Keyword(s):  
Machine Learning ◽  
De Novo ◽  
Current Reference ◽  
Marine Snails ◽  
Rnaseq Data ◽  
Venom Glands ◽  
Reference Databases ◽  
Conus Venom ◽  
Chemical Characters ◽  
Drug Leads

Cone snails (genus Conus) are venomous marine snails that inject prey with a lethal cocktail of conotoxins, small, secreted, and cysteine-rich peptides. Given the diversity and often high affinity for their molecular targets, consisting of ion channels, receptors or transporters, many conotoxins have become invaluable pharmacological probes, drug leads, and therapeutics. Transcriptome sequencing of Conus venom glands followed by de novo assembly and homology-based toxin identification and annotation is currently the state-of-the-art for discovery of new conotoxins. However, homology-based search techniques, by definition, can only detect novel toxins that are homologous to previously reported conotoxins. To overcome these obstacles for discovery, we have created ConusPipe, a machine learning tool that utilizes prominent chemical characters of conotoxins to predict whether a certain transcript in a Conus transcriptome, which has no otherwise detectable homologs in current reference databases, is a putative conotoxin. By using ConusPipe on RNASeq data of 10 species, we report 5148 new putative conotoxin transcripts that have no homologues in current reference databases. 896 of these were identified by at least three out of four models used. These data significantly expand current publicly available conotoxin datasets and our approach provides a new computational avenue for the discovery of novel toxin families.

scholarly journals Discovery of Novel Conotoxin Candidates Using Machine Learning

2018 ◽  
Author(s):  
Qing Li ◽  
Maren Watkins ◽  
Samuel D. Robinson ◽  
Helena Safavi-Hemami ◽  
Mark Yandell
Keyword(s):  
Machine Learning ◽  
De Novo ◽  
Current Reference ◽  
Marine Snails ◽  
Rnaseq Data ◽  
Venom Glands ◽  
Reference Databases ◽  
Conus Venom ◽  
Chemical Characters ◽  
Drug Leads

Cone snails (genus Conus) are venomous marine snails that inject prey with a lethal cocktail of conotoxins, small, secreted, cysteine-rich peptides. Given the diversity and often high affinity for their molecular targets, consisting of ion channels, receptors or transporters, many conotoxins have become invaluable pharmacological probes, drug leads and therapeutics. Transcriptome sequencing of Conus venom glands followed by de novo assembly and homology-based toxin identification and annotation is currently the state-of-the-art for discovery of new conotoxins. However, homology-based search techniques, by definition, can only detect novel toxins that are homologous to previously reported conotoxins. To overcome these obstacles for discovery we have created ConusPipe, a machine learning tool that utilizes prominent chemical characters of conotoxins to predict whether a certain transcript in a Conus transcriptome, which has no otherwise detectable homologs in current reference databases, is a putative conotoxin. By using ConusPipe on RNASeq data of 10 species, we report 5,230 new putative conotoxin transcripts that have no homologues in current reference databases. 893 of these were identified by at least 3 out of 4 models used. These data significantly expand current publicly available conotoxin datasets and our approach provides a new computational avenue for the discovery of novel toxin families.

scholarly journals Machine learning differentiates enzymatic and non-enzymatic metals in proteins

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Ryan Feehan ◽  
Meghan W. Franklin ◽  
Joanna S. G. Slusky
Keyword(s):  
Machine Learning ◽  
Metal Binding ◽  
Binding Sites ◽  
Active Sites ◽  
De Novo ◽  
Enzyme Design ◽  
Metal Binding Sites ◽  
Ensemble Machine Learning ◽  
Machine Learning Model ◽  
Physicochemical Features

AbstractMetalloenzymes are 40% of all enzymes and can perform all seven classes of enzyme reactions. Because of the physicochemical similarities between the active sites of metalloenzymes and inactive metal binding sites, it is challenging to differentiate between them. Yet distinguishing these two classes is critical for the identification of both native and designed enzymes. Because of similarities between catalytic and non-catalytic  metal binding sites, finding physicochemical features that distinguish these two types of metal sites can indicate aspects that are critical to enzyme function. In this work, we develop the largest structural dataset of enzymatic and non-enzymatic metalloprotein sites to date. We then use a decision-tree ensemble machine learning model to classify metals bound to proteins as enzymatic or non-enzymatic with 92.2% precision and 90.1% recall. Our model scores electrostatic and pocket lining features as more important than pocket volume, despite the fact that volume is the most quantitatively different feature between enzyme and non-enzymatic sites. Finally, we find our model has overall better performance in a side-to-side comparison against other methods that differentiate enzymatic from non-enzymatic sequences. We anticipate that our model’s ability to correctly identify which metal sites are responsible for enzymatic activity could enable identification of new enzymatic mechanisms and de novo enzyme design.

AStrap: identification of alternative splicing from transcript sequences without a reference genome

2018 ◽  
Vol 35 (15) ◽  
pp. 2654-2656 ◽  
Author(s):  
Guoli Ji ◽  
Wenbin Ye ◽  
Yaru Su ◽  
Moliang Chen ◽  
Guangzao Huang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alternative Splicing ◽  
Single Molecule ◽  
Reference Genome ◽  
De Novo ◽  
Supplementary Information ◽  
Model Organisms ◽  
Sequencing Data ◽  
Extensive Evaluation ◽  
Reference Genomes

Abstract Summary Alternative splicing (AS) is a well-established mechanism for increasing transcriptome and proteome diversity, however, detecting AS events and distinguishing among AS types in organisms without available reference genomes remains challenging. We developed a de novo approach called AStrap for AS analysis without using a reference genome. AStrap identifies AS events by extensive pair-wise alignments of transcript sequences and predicts AS types by a machine-learning model integrating more than 500 assembled features. We evaluated AStrap using collected AS events from reference genomes of rice and human as well as single-molecule real-time sequencing data from Amborella trichopoda. Results show that AStrap can identify much more AS events with comparable or higher accuracy than the competing method. AStrap also possesses a unique feature of predicting AS types, which achieves an overall accuracy of ∼0.87 for different species. Extensive evaluation of AStrap using different parameters, sample sizes and machine-learning models on different species also demonstrates the robustness and flexibility of AStrap. AStrap could be a valuable addition to the community for the study of AS in non-model organisms with limited genetic resources. Availability and implementation AStrap is available for download at https://github.com/BMILAB/AStrap. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Machine learning approaches to predict the Plant-associated phenotype of Xanthomonas strains

BMC Genomics ◽  
2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Dennie te Molder ◽  
Wasin Poncheewin ◽  
Peter J. Schaap ◽  
Jasper J. Koehorst
Keyword(s):  
Machine Learning ◽  
Plant Pathogens ◽  
De Novo ◽  
Classification Algorithms ◽  
Learning Approaches ◽  
Enabling Factors ◽  
Essential Step ◽  
The World ◽  
Genome Content

Abstract Background The genus Xanthomonas has long been considered to consist predominantly of plant pathogens, but over the last decade there has been an increasing number of reports on non-pathogenic and endophytic members. As Xanthomonas species are prevalent pathogens on a wide variety of important crops around the world, there is a need to distinguish between these plant-associated phenotypes. To date a large number of Xanthomonas genomes have been sequenced, which enables the application of machine learning (ML) approaches on the genome content to predict this phenotype. Until now such approaches to the pathogenomics of Xanthomonas strains have been hampered by the fragmentation of information regarding pathogenicity of individual strains over many studies. Unification of this information into a single resource was therefore considered to be an essential step. Results Mining of 39 papers considering both plant-associated phenotypes, allowed for a phenotypic classification of 578 Xanthomonas strains. For 65 plant-pathogenic and 53 non-pathogenic strains the corresponding genomes were available and de novo annotated for the presence of Pfam protein domains used as features to train and compare three ML classification algorithms; CART, Lasso and Random Forest. Conclusion The literature resource in combination with recursive feature extraction used in the ML classification algorithms provided further insights into the virulence enabling factors, but also highlighted domains linked to traits not present in pathogenic strains.

scholarly journals Choice of assembly software has a critical impact on virome characterisation

2018 ◽  
Author(s):  
Thomas D.S. Sutton ◽  
Adam G. Clooney ◽  
Feargal J. Ryan ◽  
R. Paul Ross ◽  
Colin Hill
Keyword(s):  
De Novo ◽  
Vital Role ◽  
Well Performance ◽  
Community Members ◽  
Reference Databases ◽  
Downstream Analysis ◽  
Metagenomic Assembly ◽  
Assembly Performance ◽  
Genomic Repeats

AbstractBackgroundThe viral component of microbial communities play a vital role in driving bacterial diversity, facilitating nutrient turnover and shaping community composition. Despite their importance, the vast majority of viral sequences are poorly annotated and share little or no homology to reference databases. As a result, investigation of the viral metagenome (virome) relies heavily on de novo assembly of short sequencing reads to recover compositional and functional information. Metagenomic assembly is particularly challenging for virome data, often resulting in fragmented assemblies and poor recovery of viral community members. Despite the essential role of assembly in virome analysis and difficulties posed by these data, current assembly comparisons have been limited to subsections of virome studies or bacterial datasets.DesignThis study presents the most comprehensive virome assembly comparison to date, featuring 16 metagenomic assembly approaches which have featured in human virome studies. Assemblers were assessed using four independent virome datasets, namely; simulated reads, two mock communities, viromes spiked with a known phage and human gut viromes.ResultsAssembly performance varied significantly across all test datasets, with SPAdes (meta) performing consistently well. Performance of MIRA and VICUNA varied, highlighting the importance of using a range of datasets when comparing assembly programs. It was also found that while some assemblers addressed the challenges of virome data better than others, all assemblers had limitations. Low read coverage and genomic repeats resulted in assemblies with poor genome recovery, high degrees of fragmentation and low accuracy contigs across all assemblers. These limitations must be considered when setting thresholds for downstream analysis and when drawing conclusions from virome data.

scholarly journals iBitter-Fuse: A Novel Sequence-Based Bitter Peptide Predictor by Fusing Multi-View Features

2021 ◽  
Vol 22 (16) ◽  
pp. 8958
Author(s):  
Phasit Charoenkwan ◽  
Chanin Nantasenamat ◽  
Md. Mehedi Hasan ◽  
Mohammad Ali Moni ◽  
Pietro Lio’ ◽  
...  
Keyword(s):  
Machine Learning ◽  
Large Scale ◽  
De Novo ◽  
Predictive Performance ◽  
Support Vector ◽  
Sufficient Information ◽  
Self Assessment ◽  
Accurate Identification ◽  
Bitter Peptides ◽  
Accurate Performance

Accurate identification of bitter peptides is of great importance for better understanding their biochemical and biophysical properties. To date, machine learning-based methods have become effective approaches for providing a good avenue for identifying potential bitter peptides from large-scale protein datasets. Although few machine learning-based predictors have been developed for identifying the bitterness of peptides, their prediction performances could be improved. In this study, we developed a new predictor (named iBitter-Fuse) for achieving more accurate identification of bitter peptides. In the proposed iBitter-Fuse, we have integrated a variety of feature encoding schemes for providing sufficient information from different aspects, namely consisting of compositional information and physicochemical properties. To enhance the predictive performance, the customized genetic algorithm utilizing self-assessment-report (GA-SAR) was employed for identifying informative features followed by inputting optimal ones into a support vector machine (SVM)-based classifier for developing the final model (iBitter-Fuse). Benchmarking experiments based on both 10-fold cross-validation and independent tests indicated that the iBitter-Fuse was able to achieve more accurate performance as compared to state-of-the-art methods. To facilitate the high-throughput identification of bitter peptides, the iBitter-Fuse web server was established and made freely available online. It is anticipated that the iBitter-Fuse will be a useful tool for aiding the discovery and de novo design of bitter peptides

scholarly journals A Program for Automated Data Mining of PubChem to Screen a Billion Compounds and Generate by Machine Learning Based AutoQSAR Algorithm Anti-Corona Viral Drug Leads (Replicase Polyprotein 1ab Inhibitors) and in Silico Study of the Top Drug Lead Compounds

2020 ◽  
Author(s):  
Ben Geoffrey A S ◽  
Akhil Sanker ◽  
Host Antony Davidd ◽  
Judith Gracia
Keyword(s):  
Machine Learning ◽  
Data Mining ◽  
In Silico ◽  
Drug Target ◽  
Data Set ◽  
Lead Compounds ◽  
Pubchem Database ◽  
Drug Lead ◽  
Lead Generation ◽  
Drug Leads

Our work is composed of a python program for automatic data mining of PubChem database to collect data associated with the corona virus drug target replicase polyprotein 1ab (UniProt identifier : POC6X7 ) of data set involving active compounds, their activity value (IC50) and their chemical/molecular descriptors to run a machine learning based AutoQSAR algorithm on the data set to generate anti-corona viral drug leads. The machine learning based AutoQSAR algorithm involves feature selection, QSAR modelling, validation and prediction. The drug leads generated each time the program is run is reflective of the constantly growing PubChem database is an important dynamic feature of the program which facilitates fast and dynamic anti-corona viral drug lead generation reflective of the constantly growing PubChem database. The program prints out the top anti-corona viral drug leads after screening PubChem library which is over a billion compounds. The interaction of top drug lead compounds generated by the program and two corona viral drug target proteins, 3-Cystiene like Protease (3CLPro) and Papain like protease (PLpro) was studied and analysed using molecular docking tools. The compounds generated as drug leads by the program showed favourable interaction with the drug target proteins and thus we recommend the program for use in anti-corona viral compound drug lead generation as it helps reduce the complexity of virtual screening and ushers in an age of automatic ease in drug lead generation. The leads generated by the program can further be tested for drug potential through further In Silico, In Vitro and In Vivo testing <div><br></div><div><div>The program is hosted, maintained and supported at the GitHub repository link given below</div><div><br></div><div>https://github.com/bengeof/Drug-Discovery-P0C6X7</div></div><div><br></div>

scholarly journals Setting the stage for the machine intelligence era in marine science

2020 ◽  
Vol 77 (4) ◽  
pp. 1267-1273
Author(s):  
Cigdem Beyan ◽  
Howard I Browman
Keyword(s):  
Machine Learning ◽  
De Novo ◽  
Machine Intelligence ◽  
Sensor Technology ◽  
Marine Science ◽  
Learning Approaches ◽  
Marine Research ◽  
And Function ◽  
Short Time ◽  
Automated Decision Making

Abstract Machine learning, a subfield of artificial intelligence, offers various methods that can be applied in marine science. It supports data-driven learning, which can result in automated decision making of de novo data. It has significant advantages compared with manual analyses that are labour intensive and require considerable time. Machine learning approaches have great potential to improve the quality and extent of marine research by identifying latent patterns and hidden trends, particularly in large datasets that are intractable using other approaches. New sensor technology supports collection of large amounts of data from the marine environment. The rapidly developing machine learning subfield known as deep learning—which applies algorithms (artificial neural networks) inspired by the structure and function of the brain—is able to solve very complex problems by processing big datasets in a short time, sometimes achieving better performance than human experts. Given the opportunities that machine learning can provide, its integration into marine science and marine resource management is inevitable. The purpose of this themed set of articles is to provide as wide a selection as possible of case studies that demonstrate the applications, utility, and promise of machine learning in marine science. We also provide a forward-look by envisioning a marine science of the future into which machine learning has been fully incorporated.

scholarly journals Machine Learning–Based Risk Assessment for Cancer Therapy–Related Cardiac Dysfunction in 4300 Longitudinal Oncology Patients

2020 ◽  
Vol 9 (23) ◽  
Author(s):  
Yadi Zhou ◽  
Yuan Hou ◽  
Muzna Hussain ◽  
Sherry‐Ann Brown ◽  
Thomas Budd ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cancer Therapy ◽  
Cancer Patients ◽  
Cardiac Dysfunction ◽  
De Novo ◽  
Left Ventricular ◽  
Left Ventricular Ejection ◽  
Oncology Patients ◽  
Relevant Variables ◽  
After Cancer

Background The growing awareness of cardiovascular toxicity from cancer therapies has led to the emerging field of cardio‐oncology, which centers on preventing, detecting, and treating patients with cardiac dysfunction before, during, or after cancer treatment. Early detection and prevention of cancer therapy–related cardiac dysfunction (CTRCD) play important roles in precision cardio‐oncology. Methods and Results This retrospective study included 4309 cancer patients between 1997 and 2018 whose laboratory tests and cardiovascular echocardiographic variables were collected from the Cleveland Clinic institutional electronic medical record database (Epic Systems). Among these patients, 1560 (36%) were diagnosed with at least 1 type of CTRCD, and 838 (19%) developed CTRCD after cancer therapy (de novo). We posited that machine learning algorithms can be implemented to predict CTRCDs in cancer patients according to clinically relevant variables. Classification models were trained and evaluated for 6 types of cardiovascular outcomes, including coronary artery disease (area under the receiver operating characteristic curve [AUROC], 0.821; 95% CI, 0.815–0.826), atrial fibrillation (AUROC, 0.787; 95% CI, 0.782–0.792), heart failure (AUROC, 0.882; 95% CI, 0.878–0.887), stroke (AUROC, 0.660; 95% CI, 0.650–0.670), myocardial infarction (AUROC, 0.807; 95% CI, 0.799–0.816), and de novo CTRCD (AUROC, 0.802; 95% CI, 0.797–0.807). Model generalizability was further confirmed using time‐split data. Model inspection revealed several clinically relevant variables significantly associated with CTRCDs, including age, hypertension, glucose levels, left ventricular ejection fraction, creatinine, and aspartate aminotransferase levels. Conclusions This study suggests that machine learning approaches offer powerful tools for cardiac risk stratification in oncology patients by utilizing large‐scale, longitudinal patient data from healthcare systems.

Sign in / Sign up

Export Citation Format

Share Document