!diversity* Alpha&Diversity:*within*sample*diversity* Sample1 & Sample2 & Sample3 & Sample4 &...

download !diversity* Alpha&Diversity:*within*sample*diversity* Sample1 & Sample2 & Sample3 & Sample4 & Marker!based*metagenomic*tutorial*

of 60

  • date post

    16-Feb-2020
  • Category

    Documents

  • view

    6
  • download

    1

Embed Size (px)

Transcript of !diversity* Alpha&Diversity:*within*sample*diversity* Sample1 & Sample2 & Sample3 & Sample4 &...

  • α-­‐diversity  metrics  

    Marker-­‐based  metagenomic  tutorial   1  

  • Alpha  Diversity:  within  sample  diversity  

    Sample  1   Sample  2   Sample  3   Sample  4  

    Marker-­‐based  metagenomic  tutorial   2  

  • Alpha  Diversity:  richness  (R)  

    Marker-­‐based  metagenomic  tutorial   3  

    SPECIES  RICHNESS  (S)  ESTIMATORS:  

    •   OTU  richness  –  count  of  different  species/OTUs  

    •   Observed  Species  –  count  of  unique  OTUs  in  each  sample   •   Chao1  index  –  esGmate  diversity  from  abundance  data  (importance  of  rare  OTUs)   •   ACE  index  …  

    Sample  1   Sample  2   Sample  3   Sample  4  

    R  =  5   R  =  2   R  =  4   R  =  5  

  • Let    S0    be  the  number  of  taxa  observed  at  least  once  in  a  sample,                  a0    the  unknown  number  of  species  present  in  the  community  but  not  observed.    

    1.  OTU  richness:  R=  S0  (no  correcGon  for  taxa  not  observed);      

    2.  Chao-­‐1  index:      assumes  that  the  number  of  observaGons  for  a  taxa  has  a  Poisson   distribuGon  and  corrects  for  variance;    

             R=  S0  +  a0    i.e.     3.  ACE  (abundance-­‐based  coverage  esGmators):      involves  an  arbitrary  abundance  threshold   to  label  Sabun  as  the  number  of  abundant  taxa,  Srare  as  the  number  of  rare  taxa;  The   expression  basically  inflates  the  number  of  rare  taxa  and  inflates  again  the  number  of  taxa   with  abundance  1.    

                   R=  S0  +  a0    i.e.      

    Marker-­‐based  metagenomic  tutorial   4  

    Species  richness  indices  

  • 10 0

    15 0

    20 0

    25 0

    30 0

    35 0

    40 0

    Ri ch ne ss

    RU -B O L1 -1

    RU -K O R1

    -1

    G B- EL 75 -6 9

    FI -N -4 7

    FI -X in b- 3

    FI -F HS

    2- 11

    FI -F UT

    1- 2

    SE -G 1- 9

    FI -F SP

    1- 16

    BE -O M -2

    RU -R M 1- 9

    DE -K 35 -M u1 1

    BY -G 1- 9

    CH -H -1 49

    DE -K 35 -Ii nb

    KE -1 -1

    RU -Y AK

    1- 1

    FR -C 1- 1

    IT -IS

    R1 -8

    IL -M 1- 8

    IR -G G 1- 7

    Marker-­‐based  metagenomic  tutorial   5  

    Species  richness:  example  of  use  

    Daphnia  clones  

    Ri ch ne

    ss  (o

    bs er ve d   sp ec ie s)   Influence  of  temperature  on  Daphnia  microbiota:  

      20°C    

    28°C  

  • SPECIES  RICHNESS  ESTIMATORS     RICHNESS  and  EVENNESS  ESTIMATORS:  the  calculated  value  of  diversity  increases   both  when  the  number  of  species  increases  and  when  evenness  increases.    

    •   InformaSon  staSsScs:  Shannon-­‐Wiener,  Shannon-­‐Weaver,  Shannon  entropy  

    •   Dominance  indices:  Inverse  Simpson,  Gini–Simpson,  Berger–Parker  index  

    Alpha  Diversity:  within  sample  diversity  

    Marker-­‐based  metagenomic  tutorial   6  

    Sample  1   Sample  3   Sample  4  

    R  =  5  

    Sample  2  

    R  =  2   R  =  4   R  =  5  

  •  Refers  to  how  common  or  rare  a  species  is  relaGve  to  other  species  in  a  community.  

    Alpha  Diversity:  relaGve  abundances  

    Marker-­‐based  metagenomic  tutorial   7  

    Sample  1   Sample  3   Sample  4  Sample  2  

    RelaGve  species  abundance  distribuGons  are  graphed  as    rank-­‐abundance  diagrams.  

  • Alpha  Diversity:  species  evenness  

    Marker-­‐based  metagenomic  tutorial   8  

     Species  evenness  refers  to  how  equally  abundant  species  in  an  environment  are.  

    Increasing  richness  and  abundance  

    Sample  1  Sample  2   Sample  3   Sample  4  

  • Alpha  Diversity:  indices’  performance  

    Sample size sensitivity!

    Influenced  by  rare  OTUs   Influenced  by  dominance/abundance  of    OTUs  

    Marker-­‐based  metagenomic  tutorial   9  

    Shannon"

    Species richness"

    Diversity Index Value!

    Inverse Simpson"

    Berger-Parker"

    Sequencing depth!

  • SPECIES  RICHNESS  ESTIMATORS     RICHNESS  and  EVENNESS  ESTIMATORS     PHYLOGENETIC  RICHNESS  ESTIMATOR:    

    •   PhylogeneSc  diversity  (PD)  –  takes  into  consideraGon  the  phylogeny  of   microbes  to  esGmate  diversity  across  a  tree  

    Alpha  Diversity:  phylogeneGc  diversity  

    Marker-­‐based  metagenomic  tutorial   10  

    R  =  5  

    Sample  1   Sample  3   Sample  4  Sample  2  

    R  =  2   R  =  4   R  =  5  

  • Marker-­‐based  metagenomic  tutorial   11  

    Alpha  Diversity:  phylogeneGc  diversity  

    Sample  2  

    R  =  2  

    OTU  2  OTU  1  

  • Marker-­‐based  metagenomic  tutorial   12  

    Alpha  Diversity:  phylogeneGc  diversity  

    Sample  2  

    R  =  2  

    OTU  2  OTU  1  

    Phylogene/c  diversity  is  high  

  • Marker-­‐based  metagenomic  tutorial   13  

    Alpha  Diversity:  phylogeneGc  diversity  

    Sample  2  

    R  =  2  

    OTU  2  OTU  1  

  • Marker-­‐based  metagenomic  tutorial   14  

    Alpha  Diversity:  phylogeneGc  diversity  

    Sample  2  

    R  =  2  

    OTU  2  OTU  1  

    Phylogene/c  diversity  is  low  

  • »  What  you  want  to  know  determines  how  you   analyze  your  data  

    »  How  important  is  each  aspect  of  diversity?   ˃  Richness?     ˃  Evenness?     ˃  Dominance?   ˃  Abundance?     ˃  Per-­‐species  (relaGve)  abundance?   ˃  Taxon  diversity?  

    Marker-­‐based  metagenomic  tutorial   15  

  • VariaGon  among  samples  

    Marker-­‐based  metagenomic  tutorial   16  

  • Shannon"

    Species richness"

    Diversity Index Value!

    Inverse Simpson"

    Berger-Parker"

    Sequencing depth!

    Diversity:  influence  of  sequencing  effort  

    Marker-­‐based  metagenomic  tutorial   17  

  • Marker-­‐based  metagenomic  tutorial   18  

    VariaSon  in  reads  counts  

  • Sample Addition Sequence! R

    ic hn

    es s!

    Samples: Accumulation"

    Samples: Rarefaction"

    Taxa: Accumulation" Taxa: Rarefaction"

    RarefacSon  

    Marker-­‐based  metagenomic  tutorial   19  

    Rarefying  was  first  recommended  for   microbiome  counts  in  order  to  moderate   differences  in  the  presence  of  rare  OTUs    

    (Lozupone  et  al.  2011  ISME  J)     Goal:    

    •  Standardize  unequal  sequencing  effort.   •  Enable  similarity  comparisons  along  a  

    range  of  samples  or  a  gradient.   •  Enable  comparison  of  different  runs  or  

    replicates.  

    Procedure:    

    •  Determine  the  minimum  sequencing  depth.   •  Subsample  without  replacement  sequences  from  the  larger  libraries  so  that  all  

    have  the  same  smallest  size.     •  Note  that  the  term  is  a  bit  misleading  as  this  step  should  really  be  called  

    "subsampling  to  a  given