Machine Learning for detection of viral sequences in human metagenomic datasets

BackgroundDetection of highly divergent or yet unknown viruses from metagenomics sequencing datasets is a major bioinformatics challenge. When human samples are sequenced, a large proportion of assembled contigs are classified as “unknown”, as conventional methods find no similarity to known sequences. We wished to explore whether machine learning algorithms using Relative Synonymous Codon Usage frequency (RSCU) could improve the detection of viral sequences in metagenomic sequencing data.ResultsWe trained Random Forest and Artificial Neural Network using metagenomic sequences taxonomically classified into virus and non-virus classes. The algorithms achieved accuracies well beyond chance level, with area under ROC curve 0.79. Two codons (TCG and CGC) were found to have a particularly strong discriminative capacity.ConclusionRSCU-based machine learning techniques applied to metagenomic sequencing data can help identify a large number of putative viral sequences and provide an addition to conventional methods for taxonomic classification.

[1]  Piotr Bala,et al.  Massively Parallel Implementation of Sequence Alignment with Basic Local Alignment Search Tool Using Parallel Computing in Java Library , 2018, J. Comput. Biol..

[2]  N. Friedman,et al.  Trinity : reconstructing a full-length transcriptome without a genome from RNA-Seq data , 2016 .

[3]  Forest Rohwer,et al.  Metagenomic Analysis of Respiratory Tract DNA Viral Communities in Cystic Fibrosis and Non-Cystic Fibrosis Individuals , 2009, PloS one.

[4]  Vincent Van Asch,et al.  Macro-and micro-averaged evaluation measures [ [ BASIC DRAFT ] ] , 2013 .

[5]  Laurie Goodman,et al.  Large and linked in scientific publishing , 2012, GigaScience.

[6]  Zhengwei Zhu,et al.  CD-HIT: accelerated for clustering the next-generation sequencing data , 2012, Bioinform..

[7]  J. Gilbert,et al.  Metagenomics - a guide from sampling to data analysis , 2012, Microbial Informatics and Experimentation.

[8]  Joakim Dillner,et al.  Extension of the viral ecology in humans using viral profile hidden Markov models , 2018, PloS one.

[9]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[10]  Kellie J. Archer,et al.  Empirical characterization of random forest variable importance measures , 2008, Comput. Stat. Data Anal..

[11]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[12]  NowickiMarek,et al.  Massively Parallel Implementation of Sequence Alignment with Basic Local Alignment Search Tool Using Parallel Computing in Java Library. , 2018 .

[13]  Paul M. Sharp,et al.  Codon usage in yeast: cluster analysis clearly differentiates highly and lowly expressed genes , 1986, Nucleic Acids Res..

[14]  N. Friedman,et al.  Trinity: reconstructing a full-length transcriptome without a genome from RNA-Seq data , 2011, Nature Biotechnology.

[15]  Luis V. Santana-Quintero,et al.  A new and updated resource for codon usage tables , 2017, BMC Bioinformatics.

[16]  J. Moodley,et al.  Next-generation sequencing of cervical DNA detects human papillomavirus types not detected by commercial kits , 2012, Virology Journal.

[17]  German Tischler,et al.  Next-generation sequencing and large genome assemblies. , 2012, Pharmacogenomics.

[18]  George M. Weinstock,et al.  Sequence Analysis of the Human Virome in Febrile and Afebrile Children , 2012, PloS one.

[19]  Siu-Ming Yiu,et al.  IDBA-UD: a de novo assembler for single-cell and metagenomic sequencing data with highly uneven depth , 2012, Bioinform..

[20]  G. Weinstock,et al.  Emerging view of the human virome , 2012, Translational Research.

[21]  Joakim Dillner,et al.  Viremia during pregnancy and risk of childhood leukemia and lymphomas in the offspring: Nested case–control study , 2016, International journal of cancer.

[22]  J. Derisi,et al.  Profile Hidden Markov Models for the Detection of Viruses within Metagenomic Sequence Data , 2014, PloS one.

[23]  Joakim Dillner,et al.  Does human papillomavirus-negative condylomata exist? , 2015, Virology.

[24]  Joakim Dillner,et al.  Human papillomavirus type 197 is commonly present in skin tumors , 2015, International journal of cancer.

[25]  R. Desrosiers,et al.  Importance of codon usage for the temporal regulation of viral gene expression , 2015, Proceedings of the National Academy of Sciences.

[26]  Joakim Dillner,et al.  Metagenomic sequencing of "HPV-negative" condylomas detects novel putative HPV types. , 2013, Virology.

[27]  Joakim Dillner,et al.  Phylogenetically diverse TT virus viremia among pregnant women. , 2012, Virology.

[28]  Siu-Ming Yiu,et al.  Erratum: SOAPdenovo2: an empirically improved memory-efficient short-read de novo assembler , 2015, GigaScience.

[29]  Joakim Dillner,et al.  Deep sequencing extends the diversity of human papillomaviruses in human skin , 2014, Scientific Reports.

[30]  Richard Durbin,et al.  Fast and accurate long-read alignment with Burrows–Wheeler transform , 2010, Bioinform..

[31]  Davit Bzhalava,et al.  Massively Parallel Implementation of Sequence Alignment with BLAST Using PCJ Library , 2018 .

[32]  Joakim Dillner,et al.  Detection of DNA viruses in prostate cancer , 2016, Scientific Reports.

[33]  Curtis A Suttle,et al.  Previously unknown and highly divergent ssDNA viruses populate the oceans , 2013, The ISME Journal.

[34]  Joakim Dillner,et al.  Bioinformatics for Viral Metagenomics , 2013 .

[35]  Richard Durbin,et al.  Sequence analysis Fast and accurate short read alignment with Burrows – Wheeler transform , 2009 .

[36]  P. Moore,et al.  Clonal Integration of a Polyomavirus in Human Merkel Cell Carcinoma , 2008, Science.

[37]  Marc Eloit,et al.  The human virome: new tools and concepts , 2013, Trends in Microbiology.

[38]  O. Dereure,et al.  Human Skin Microbiota: High Diversity of DNA Viruses Identified on the Human Skin by High Throughput Sequencing , 2012, PloS one.

[39]  Fernando Castro-Chavez,et al.  Most Used Codons per Amino Acid and per Genome in the Code of Man Compared to Other Organisms According to the Rotating Circular Genetic Code. , 2011, NeuroQuantology : an interdisciplinary journal of neuroscience and quantum physics.

[40]  Joakim Dillner,et al.  High throughput sequencing reveals diversity of Human Papillomaviruses in cutaneous lesions , 2011, International journal of cancer.

[41]  Achim Zeileis,et al.  BMC Bioinformatics BioMed Central Methodology article Conditional variable importance for random forests , 2008 .

[42]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[43]  Jian Wang,et al.  SOAPdenovo2: an empirically improved memory-efficient short-read de novo assembler , 2012, GigaScience.

[44]  J. Paoli,et al.  Unbiased Approach for Virus Detection in Skin Lesions , 2013, PloS one.

[45]  Forest Rohwer,et al.  Case studies of the spatial heterogeneity of DNA viruses in the cystic fibrosis lung. , 2012, American journal of respiratory cell and molecular biology.

[46]  Haomiao Jia,et al.  Discerning suicide in drug intoxication deaths: Paucity and primacy of suicide notes and psychiatric history , 2018, PloS one.

[47]  W. Ian Lipkin,et al.  Newly Discovered Ebola Virus Associated with Hemorrhagic Fever Outbreak in Uganda , 2008, PLoS pathogens.