Microbiome Robert Kraaij, PhD Erasmus MC, Internal Medicine - PowerPoint PPT Presentation

SNPs and Human Diseases XV November 14th, 2018 Microbiome Robert Kraaij, PhD Erasmus MC, Internal Medicine r.kraaij@erasmusmc.nl

Metagenomics - terminology  the study of metagenomes  genetic material recovered from environmental samples  ecological community of microorganisms  symbiosis  commensal  mutual  parasitic OPTION 1:  microbiota  community of microorganisms  microbiome  genomes of the microbiota

Metagenomics - terminology  the study of metagenomes  genetic material recovered from environmental samples  ecological community of microorganisms  symbiosis  commensal  mutual  parasitic OPTION 2:  microbiota  collection of microorganisms  metagenome  genomes of the microbiota  microbiome  community of microorganisms and host

Microbiota: more than just bacteria…  Archaea  Bacteria  Protozoa  Viruses  human viruses  bacteriophages  Fungi  molds  yeasts

Microbiota: more than just bacteria…  Archaea  Bacteria  Protozoa  Viruses  human viruses  bacteriophages  Fungi  molds  yeasts

The human gut microbiota - the forgotten organ  10 13 bacterial cells = 10 13 body cells  ~10 6 bacterial genes vs ~20,000 human genes  many unique functions  involved in health and disease!

Density of microbiota increases along GI tract stool (10 11 cells/ml) Walter and Ley (2011) Annu Rev Microbiol.

Stool as ‘proxy’ of gut ( distal colon) microbiota profiling collection storage metadata - Bristol stool scale - Rotterdam Study RS-IV - n = 836 type 1 type 2 type 3 type 4 type 5 type 6 type 7

Human microbiota: more than just the gut… eye skin nose stool tooth urine

Microbiota GENOTYPE MICROBIOME ENVIRONMENT PHENOTYPE DIET LIFE-STYLE

Gut microbiome and disease associations  obesity  Crohn’s disease hype cycle  ulcerative colitis  eczema  asthma  diabetes  depression  etc

Overview  Microbiota profiling  Data analysis

Microbiota profiling  W HO ARE THEY ?  W HAT DO THEY DO ?

Microbiota profiling  culture-based techniques  culturomics  16S rRNA marker gene  arrays (hitChip)  ISpro  sequencing  microbiome array  shotgun sequencing (metagenomics)

IS-pro TM profiling  16S-23S interspace (IS region)  taxonomy based on size differences prokaryotic 16S 23S 5S rRNA operon IS region Bacteroidetes Firmicutes , Actinobacteria , FAFV Fusobacteria , Verrucomicrobia Proteobacteria abundance NCBI database - 16S – 23S rRNA - 8990 entries fragment size (nt) Budding et al . (2010) FASEB J.

Microbiota profiling  culture-based techniques  culturomics  16S rRNA marker gene  arrays (hitChip)  ISpro  sequencing  microbiome array  shotgun sequencing (metagenomics)

16S ribosomal RNA gene amplicon  highly conserved in bacteria and archaea  species-independent PCR amplification  variable regions  taxonomic classification 16S rRNA

16S rRNA amplicons prokaryotic 16S 23S 5S rRNA operon IS region 1500bp  Oxford Nanopore long read sequencing ~400bp  Illumina MiSeq short read sequencing

16S RNA analysis pipeline DNA isolation 16S rRNA Sequencing Analysis (NorDiag Arrow) amplicon (Illumina MiSeq) (QIIME)

16S rRNA amplicon and sequencing Illumina MiSeq Fadrosh et al. (2014)

QIIME-based analysis pipeline Read-pair merging Q-score > 19 Sample QC Reads > mean – 2 SD Chimera filtering Silva database, version 128 Max Planck Institute for Marine Microbiology and Jacobs University, Bremen, Germany OTU calling  September 2016 Taxonomy Phylogeny  8,430,487 entries OTU abundancy filtering > 0.005% of total reads OTU table (anonymous) Biome table (taxonomy) Phylogenetic tree Caporaso et al. (2010)

Read-pair merging

Chimera filtering - chimeras are PCR artifacts

Chimera filtering Query Chunk Chunk Chunk Chunk Ref DB Hits normal chimera A 4x A Query Query B

OTU clustering Operational taxonomic units (OTUs)  clustering on basis of homology of the reads (97%)  OTUs can be aligned to reference databases  unknown OTUs can still be used in analyses

Closed reference calling  Each read is compared directly to the database  Database determines phylogenetic tree  Standardized taxonomy > allows for collaboration

Microbiota profiling  culture-based techniques  culturomics  16S rRNA marker gene  arrays (hitChip)  ISpro  sequencing  microbiome array  shotgun sequencing (metagenomics)

Affymetrix Axiom Microbiome array

Microbiota profiling  culture-based techniques  culturomics  16S rRNA marker gene  arrays (hitChip)  ISpro  sequencing  microbiome array  shotgun sequencing (metagenomics)

Shotgun metagenomics Flaws of 16S rRNA profiling  selection introduced by PCR amplification  no eukaryotic species such as fungi  phylotyping will not give insights into the gene functions of unknown species

Shotgun metagenomics Direct sequencing of DNA

High output sequencing  2 x 100 bp  reads are too short for proper annotation  de novo assembly is preferred  need for compute power 2 x 100 bp ~1 kbp contigs de novo assembly paired-reads

Metagenomics technology push MetaHIT  European FP7 project Human Microbiome Project (HMP)  NIH-sponsored project

Profiling of shotgun data  phylotyping databases  metagenomic species (MGS)  ~7000 MGS specified  gene catalogue  8.1 million genes from 760 samples  functional databases

Phylotyping of shotgun data Arumugam et al., 2011  MetaHIT

Functional analysis of shotgun data Arumugam et al., 2011  MetaHIT

Taxonomic vs functional profiling  large taxonomic differences are not reflected in functional profiles Samples ordered by taxonomic profiles Samples ordered by functional profiles The Human Microbiome Project Consortium (2012)

Profiling the gut microbiome  W HO ARE THEY ?  16S TAXONOMY  M ETAGENOMICS  W HAT CAN THEY DO ?  M ETAGENOMICS  W HAT ARE THEY DOING ?  M ETATRANSCRIPTOMICS  M ETAPROTEOMICS  W HAT HAVE THEY DONE ?  M ETABOLOMICS

Profiling the gut microbiome

Overview  Microbiota profiling  Data analysis

Complex multi-dimensional data  no normal or mean profile  enterotypes?  sparse data  many zero abundances  limited by technique  count data  dependent on technique  how to normalize?  compositional data  relative abundances add up to 1

Diversities  α -diversity  diversity within a sample  biological metric  number of species * evenness  β -diversity  diversity (distance or dissimilarity) between samples  UniFrac distances

OTU table … OTU id sample_01 sample_02 sample_03 sample_04 OTU_12 3 0 456 343 OTU_318 34 45 3 2 OTU_37 567 2134 478 675 … … … … … … Total 5,975 4,952 6,735 5,374

Rotterdam 16S rRNA datasets Rotterdam Study Generation R Study adults 9-11 year-olds Domain Phylum Class Order Family Genus OTUs Domain Phylum Class Order Family Genus OTUs (1) (7) (15) (19) (36) (152) (661) (2) (11) (18) (24) (43) (183) (777) N=156 N=1,106 N=2,111 N=1,427 5 major phyla Shannon Diversity Index Shannon Diversity Index N=1,135 Radjabzadeh et al . (2018) in preparation

Children vs adults - Generation R Study vs Rotterdam Study *** 8 Shannon diversity index 7 6 5 N=2,111 N=1,427 4 3 2 GenR RS Shannon alpha diversity average phylum-level profiles Radjabzadeh et al . (2018) in preparation

MiBioGen consortium  Meta-analyses of gut microbiome GWAS Traits  > 20 cohorts (still including)  Shannon alpha-diversity  > 20,000 samples  Binary trait (presence/absence)  16S rRNA profiling (Illumina)  Quantitative trait (abundance)  226 genera  Beta-diversity  8M HRC1.1 imputed SNPs NGRC