MICROBIOLOGY JOURNAL CLUB
INFLUENCE OF SEX, HANDEDNESS, AND WASHING ON THE
DIVERSITY OF HAND SURFACE BACTERIA
The influence of sex, handedness, and washing
on the diversity of hand surface bacteria
Noah Fierera,b,1, Micah Hamadyc, Christian L. Lauberb, and Rob Knightd
aDepartment
of Ecology and Evolutionary Biology, University of Colorado, UCB 334, Boulder, CO 80309; bCooperative Institute for Research in
Environmental Sciences, University of Colorado, UCB 216, Boulder, CO 80309; cDepartment of Computer Science, University of Colorado, UCB 430,
Boulder, CO 80309; and dDepartment of Chemistry and Biochemistry, University of Colorado, UCB 215, Boulder, CO 80309
Edited by Jeffrey I. Gordon, Washington University School of Medicine, St. Louis, MO, and approved September 23, 2008 (received for review
August 11, 2008)
Bacteria thrive on and within the human body. One of the largest
human-associated microbial habitats is the skin surface, which
harbors large numbers of bacteria that can have important effects
on health. We examined the palmar surfaces of the dominant and
nondominant hands of 51 healthy young adult volunteers to
characterize bacterial diversity on hands and to assess its variability within and between individuals. We used a novel pyrosequencing-based method that allowed us to survey hand surface bacterial
communities at an unprecedented level of detail. The diversity of
skin-associated bacterial communities was surprisingly high; a
typical hand surface harbored >150 unique species-level bacterial
phylotypes, and we identified a total of 4,742 unique phylotypes
across all of the hands examined. Although there was a core set of
bacterial taxa commonly found on the palm surface, we observed
pronounced intra- and interpersonal variation in bacterial community composition: hands from the same individual shared only 17%
of their phylotypes, with different individuals sharing only 13%.
Women had significantly higher diversity than men, and community composition was significantly affected by handedness, time
since last hand washing, and an individual’s sex. The variation
within and between individuals in microbial ecology illustrated by
this study emphasizes the challenges inherent in defining what
constitutes a ‘‘healthy’’ bacterial community; addressing these
challenges will be critical for the International Human Microbiome
Project.
frequency of perturbations caused by hand washing. In addition,
pathogens may inhabit the palmar surface, and efforts to reduce
disease transmission by hand washing are a key public health
concern (9–11).
We surveyed the bacterial communities found on the palm
surfaces of both the dominant and nondominant hands of 51
undergraduate students sampled after taking an examination.
Our goal was to assess the intra- and interindividual variability
in skin-associated bacterial communities and determine how
specific factors (including sex, handedness, and time since last
hand washing) may inf luence the diversity and composition of
the bacterial communities. The 16S rRNA genes from the
palmar surface bacteria were PCR-amplified by using a universal bacterial primer set with a unique error-correcting
barcode for each sample, allowing us to analyze all of the
amplified samples in a single pyrosequencing run (12). We
extended this technique using Golay codes, which provide a
greater degree of error correction than the Hamming codes
used in the previous study, allowing us to correct any triple-bit
error and detect any quadruple-bit error (versus single-bit
correction and double-bit detection in the Hamming codes).
Coupling this barcoding technique with the high-throughput
capabilities of pyrosequencing, we were able to survey the
bacterial communities on each of the swabbed hands at an
unprecedented level of detail.
human microbiome ! pyrosequencing ! skin bacteria
Results and Discussion
After removing sequences of insufficient quality and sequences
that could not be adequately classified, nearly 332,000 sequences
acteria thrive on and within the human body, with recent
THE QUESTION
WHAT IS THE “NORMAL” FLORA OF HEALTHY HUMAN HANDS, AND
WHAT DOES VARIATION IN THIS FLORA LOOK LIKE?
PERSPECTIVE
HUMANS ARE BETTER VIEWED AS AN ECOSYSTEM
RATHER THAN A SINGLE ORGANISM
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
DGGE
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
DGGE
T-RFLP
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
DGGE
T-RFLP
FISH
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
DGGE
T-RFLP
FISH
PHYLOGENETIC ARRAYS
HOW CAN WE TAKE A CENSUS OF
MICROBIAL DIVERSITY?
CULTIVATION
RRNA SURVEYS
DGGE
T-RFLP
FISH
PHYLOGENETIC ARRAYS
454 “DEEP SEQUENCING”
Secondary Structure: small subunit ribosomal RNA
A A
U
G
U
U
G
C
G
C 1100
G
C
U
GC
C AA
700
A G
A
C
U
A
G
UG AG
U
A CGA GC
CCUUA
UCCUUUGU
CC CGG UC
CC
G
G
G UG
A
G
A A UGC
C
A UCUGGA GGA A U
G GC G
A
A
UGCUCG
A
G
G
A
A
A
C
GG GCC GG
G
G
G
G
U
U
G
A
G
UC AA
A
G
A
A
U
U
UGGA CCUU
GA
G CG G
U GGCG
G
AA
A U G
1150
G
U
G
A
U
C
G
A
G
C
U
A
A
G
A
G
C
G
C
A
GU
G
C
C
G
G
C
G
C
U
U
G
U
A GA
G
C
G
C
A
G
U
U
A
G
U
C
U
A 1200 G
A
U
A
G
C
G
C
G
U C
G
A
C
U
A
A
G
G
UC
U
G
G
C
C
G
A
U
A
C
G
G
C
G
A
U
CU
G
C
C
G
A
U
U
A
G
A
C
G
A
G
A
C
G
A
A
U
A
U
CG
U
A
U
G 800
A
A
G
A
U
G
A
G
A
A
U
C
G
U
G
U
A
G
2 ]
C
1050 C
[
U
A
m
G
G
C
A
A
A
G
G
C 750
G
C
C
G
G A C U U U U G AA G
U A
U
A
U
G
U
G
AG
G
C
G
C
G
G
G
G
C
A
A
GG
650 U
C
C
U
C A
U
A
UA
U
A C
G
C
C
1000
U
UGCA UCUGA CU GGCA A G C
A
U
A
C
A A CCUGGGA
A U
G
C
C
GC
C
C UU
G
C
U U
C
G
A
U C GGGCCCC
C
G
C
U A
GU GU A GA C U GA U U GU U U GG
C
G
C C GU
C
AG G
C G
C
A
U
A
A
A
G
G
G
G
C
A
G
A A
U
600
G
A
G
U
A
U
A
G
AC
U
C
G A
U
C C UA C
C
G
A
G
GC
U
G
A
U
A
C
G
C
G
C
G
A
G
G
G
G
G
850
A
G
C
C
C
G
UA
G
U
G
A
G
C CG
U
U
GU C GA CUU
A
U
GCC
A
A U AC
U
A
C
ACC
G
A
U A UU G
C
A
A
C
C
G
G
G
A
G
A
C
U
m7
A
A
A
G
C
U
U
C
A
G
C
U
G
C
C
G
G
C
C
G
UG
A
A G
U
UG
C
G
A
A
A
U
A
C
G
CG
U
A
G
A
C
G
1250
G A
U
G
A
A
G
C
A
C
G
A
G
A
U
C
C
G
C
A
GC
U
5
A
C
950 U
G
U
U
AAAG
G
mC 2 U
C
U
G
mG
G G
CA U
A
U
A
C
C
C
A
G
U
U
U
U C AG
CA
C
G
A
A
A
C
C
A G UA
G
G
G
G
C
G
C
U
A
U
A
A
ψ
G
A
A
G
G
G
G
G
C
C
A
U
G
G
U
G
C
U
C U
C
C
G
A
U
G
G
A
G
C
G
U
C
A 900
U A A
A A G
A
G
CA A
C
G
G
CUC
UUGG A
U GG
A
C
C
AAAA
G
U
A
A
C
G
U
A
450
AA
U
A
C
G
G
UA A C
G G
G
C
G
G
UG
C
G
G
C
C
G
G
A A
A
G
G
A CG G C
U
U UG A CG GGGCCCG A CA A
G
C
U
C A
G
C
G
G
G
G
U
A 500 C
A
CG UACU
C
A
CU
A
U 550 A A
A
G
U
G
C
A C UGU UCCGGGC UG
GA
G C
G
A
A
U
A
G
U 1300
U
U
C A
C U C
U
GU U G
U
A G
U
G
C
C
A
C
UAA
C
G
A
U
G
A
U
C
G
A
A
U G
U
A C
A
G
G
A
U
G GGUU
GUA C
G
A
A
C
C
A
A
G
A
A 10
C
A 1350
G
A
A
A
G
U
1400 C
C
GU
A A
U
U A U G U GGC
U CCGG
G
U
U
U
C
C
A
G 400
G
C
U
G
4G
U
U
AGAA
G
C
G G
C
A
G
m
m
A
G
G
A
C
U
AU
G
C
3A
C
C
G
m
G
G
A
U
G
A
C
m2G
C
C
G
G
A
C
C
A
C
1500 U A A C C G U A G G
A
U
A GCCUG A UGC A G
A
G
G
C
G
C
U
G
A
C
U
U
U
G
G
m6A
C
A
5
2
C
GUUGGCGUCC 6
GA
C
5’
mC
U
A C G UA
G CGGGU
U
A
G
mA
A
G
2
A A G
U U
A
AAC
G
C
C
G
A
A
C
A
U
U
U
A
CA A
C
G
C
G
C
G
A
C
G
C
A
G
C
UA 50
G
C
G
U
U
U
G
G
CA
C
C
G
A
U
C
G
A
CA
G
U
A
G
U
A
C G
U
G
U
A
C
G
G 350
A
U
A
G
C
A G
U
A GC
G
C
C
G
A
C
U
G
3’
U
G
U
C
A
CG
A
U
A
C
GA
AA
A
GUC
A
U
G
CAG
GUC A CG GU CA G GA A GA A GC
G
G
U
U
300 A
G
C
U
GU C A A G G U
CGG UGA CA GUC UUUCUUCG
A
U
C
A
G G
G
G
A
U
GA G A
A
G
A
A
U
U C
C
G
G
100
CA
A
A
UC
C
GC C
A
UG
G
G
A
G G C
U
A
GU
G
C
CG
A
A
G
C
A
G
U
U
A
U
C
A
U
A
C
A
G
U
C
GU
GU
G
C
G A
CG AU
U
G
U
G
A
G
C
A
G
U UA
G
C
A
U
C
C
G
C
G
G
A
U
G
G
C
G
G
U
G
C
C
G
A
A
U
U
G
A
A
A
A
C
A
C
G
UG
C
U
A
A
C
G
A
250
G
G
C
C
G
U
G
1450 U
U
U
U
U
C
C
G
G
G
A
G
G
C
G
G
C
C
G
G
U
A
A U
C
G
U
A
150
A
U
C
G
C
G
AUA
ACUACUGG
A G G G GG
G GG CCUCUU G
A
C
G
C
- Canonical base pair (A-U, G-C)
A
U
CGCC
CGA UGGC A
U CC GGGGA G
A
A
G
U
- G-U base pair
A
AG
U A AU
U
200 A
A
G
A
- G-A base pair
A
A
C
C
C
G
U
U
- Non-canonical base pair
A
U
G
C
G
A
Every 10th nucleotide is marked with a tick mark,
A C
II
III
I
Escherichia coli
(J01695)
1. Bacteria 2. Proteobacteria 3. gamma subdivision
4. Enterobacteriaceae and related symbionts
5. Enterobacteriaceae 6. Escherichia
Nov 1999
Symbols Used In This Diagram:
and every 50th nucleotide is numbered.
Tertiary interactions with strong comparative data are connected by
solid lines.
Citation and related information available at http://www.rna.icmb.utexas.edu
Secondary Structure: small subunit ribosomal RNA
A A
U
G
U
U
G
C
G
C 1100
G
C
U
GC
C AA
700
A G
A
C
U
A
G
UG AG
U
A CGA GC
CCUUA
UCCUUUGU
CC CGG UC
CC
G
G
G UG
A
G
A A UGC
C
A UCUGGA GGA A U
G GC G
A
A
UGCUCG
A
G
G
A
A
A
C
GG GCC GG
G
G
G
G
U
U
G
A
G
UC AA
A
G
A
A
U
U
UGGA CCUU
GA
G CG G
U GGCG
G
AA
A U G
1150
G
U
G
A
U
C
G
A
G
C
U
A
A
G
A
G
C
G
C
A
GU
G
C
C
G
G
C
G
C
U
U
G
U
A GA
G
C
G
C
A
G
U
U
A
G
U
C
U
A 1200 G
A
U
A
G
C
G
C
G
U C
G
A
C
U
A
A
G
G
UC
U
G
G
C
C
G
A
U
A
C
G
G
C
G
A
U
CU
G
C
C
G
A
U
U
A
G
A
C
G
A
G
A
C
G
A
A
U
A
U
CG
U
A
U
G 800
A
A
G
A
U
G
A
G
A
A
U
C
G
U
G
U
A
G
2 ]
C
1050 C
[
U
A
m
G
G
C
A
A
A
G
G
C 750
G
C
C
G
G A C U U U U G AA G
U A
U
A
U
G
U
G
AG
G
C
G
C
G
G
G
G
C
A
A
GG
650 U
C
C
U
C A
U
A
UA
U
A C
G
C
C
1000
U
UGCA UCUGA CU GGCA A G C
A
U
A
C
A A CCUGGGA
A U
G
C
C
GC
C
C UU
G
C
U U
C
G
A
U C GGGCCCC
C
G
C
U A
GU GU A GA C U GA U U GU U U GG
C
G
C C GU
C
AG G
C G
C
A
U
A
A
A
G
G
G
G
C
A
G
A A
U
600
G
A
G
U
A
U
A
G
AC
U
C
G A
U
C C UA C
C
G
A
G
GC
U
G
A
U
A
C
G
C
G
C
G
A
G
G
G
G
G
850
A
G
C
C
C
G
UA
G
U
G
A
G
C CG
U
U
GU C GA CUU
A
U
GCC
A
A U AC
U
A
C
ACC
G
A
U A UU G
C
A
A
C
C
G
G
G
A
G
A
C
U
m7
A
A
A
G
C
U
U
C
A
G
C
U
G
C
C
G
G
C
C
G
UG
A
A G
U
UG
C
G
A
A
A
U
A
C
G
CG
U
A
G
A
C
G
1250
G A
U
G
A
A
G
C
A
C
G
A
G
A
U
C
C
G
C
A
GC
U
5
A
C
950 U
G
U
U
AAAG
G
mC 2 U
C
U
G
mG
G G
CA U
A
U
A
C
C
C
A
G
U
U
U
U C AG
CA
C
G
A
A
A
C
C
A G UA
G
G
G
G
C
G
C
U
A
U
A
A
ψ
G
A
A
G
G
G
G
G
C
C
A
U
G
G
U
G
C
U
C U
C
C
G
A
U
G
G
A
G
C
G
U
C
A 900
U A A
A A G
A
G
CA A
C
G
G
CUC
UUGG A
U GG
A
C
C
AAAA
G
U
A
A
C
G
U
A
450
AA
U
A
C
G
G
UA A C
G G
G
C
G
G
UG
C
G
G
C
C
G
G
A A
A
G
G
A CG G C
U
U UG A CG GGGCCCG A CA A
G
C
U
C A
G
C
G
G
G
G
U
A 500 C
A
CG UACU
C
A
CU
A
U 550 A A
A
G
U
G
C
A C UGU UCCGGGC UG
GA
G C
G
A
A
U
A
G
U 1300
U
U
C A
C U C
U
GU U G
U
A G
U
G
C
C
A
C
UAA
C
G
A
U
G
A
U
C
G
A
A
U G
U
A C
A
G
G
A
U
G GGUU
GUA C
G
A
A
C
C
A
A
G
A
A 10
C
A 1350
G
A
A
A
G
U
1400 C
C
GU
A A
U
U A U G U GGC
U CCGG
G
U
U
U
C
C
A
G 400
G
C
U
G
4G
U
U
AGAA
G
C
G G
C
A
G
m
m
A
G
G
A
C
U
AU
G
C
3A
C
C
G
m
G
G
A
U
G
A
C
m2G
C
C
G
G
A
C
C
A
C
1500 U A A C C G U A G G
A
U
A GCCUG A UGC A G
A
G
G
C
G
C
U
G
A
C
U
U
U
G
G
m6A
C
A
5
2
C
GUUGGCGUCC 6
GA
C
5’
mC
U
A C G UA
G CGGGU
U
A
G
mA
A
G
2
A A G
U U
A
AAC
G
C
C
G
A
A
C
A
U
U
U
A
CA A
C
G
C
G
C
G
A
C
G
C
A
G
C
UA 50
G
C
G
U
U
U
G
G
CA
C
C
G
A
U
C
G
A
CA
G
U
A
G
U
A
C G
U
G
U
A
C
G
G 350
A
U
A
G
C
A G
U
A GC
G
C
C
G
A
C
U
G
3’
U
G
U
C
A
CG
A
U
A
C
GA
AA
A
GUC
A
U
G
CAG
GUC A CG GU CA G GA A GA A GC
G
G
U
U
300 A
G
C
U
GU C A A G G U
CGG UGA CA GUC UUUCUUCG
A
U
C
A
G G
G
G
A
U
GA G A
A
G
A
A
U
U C
C
G
G
100
CA
A
A
UC
C
GC C
A
UG
G
G
A
G G C
U
A
GU
G
C
CG
A
A
G
C
A
G
U
U
A
U
C
A
U
A
C
A
G
U
C
GU
GU
G
C
G A
CG AU
U
G
U
G
A
G
C
A
G
U UA
G
C
A
U
C
C
G
C
G
G
A
U
G
G
C
G
G
U
G
C
C
G
A
A
U
U
G
A
A
A
A
C
A
C
G
UG
C
U
A
A
C
G
A
250
G
G
C
C
G
U
G
1450 U
U
U
U
U
C
C
G
G
G
A
G
G
C
G
G
C
C
G
G
U
A
A U
C
G
U
A
150
A
U
C
G
C
G
AUA
ACUACUGG
A G G G GG
G GG CCUCUU G
A
C
G
C
- Canonical base pair (A-U, G-C)
A
U
CGCC
CGA UGGC A
U CC GGGGA G
A
A
G
U
- G-U base pair
A
AG
U A AU
U
200 A
A
G
A
- G-A base pair
A
A
C
C
C
G
U
U
- Non-canonical base pair
A
U
G
C
G
A
Every 10th nucleotide is marked with a tick mark,
A C
II
III
I
27F
338R
B
A
R
C
O
D
E
Escherichia coli
(J01695)
1. Bacteria 2. Proteobacteria 3. gamma subdivision
4. Enterobacteriaceae and related symbionts
5. Enterobacteriaceae 6. Escherichia
Nov 1999
Symbols Used In This Diagram:
and every 50th nucleotide is numbered.
Tertiary interactions with strong comparative data are connected by
solid lines.
Citation and related information available at http://www.rna.icmb.utexas.edu
Table S3. List of the 102 12-bp error-correcting Golay barcodes used to tag each
PCR product analyzed as part of the larger study
AACGCACGCTAG
ACACTGTTCATG
ACCAGACGATGC
ACGCTCATGGAT
ACTCACGGTATG
AGACCGTCAGAC
AGCACGAGCCTA
ACAGACCACTCA
ACCAGCGACTAG
ACGGATCGTCAG
AGCTTGACAGCT
AACTGTGCGTAC
ACCGCAGAGTCA
ACGGTGAGTGTC
ACTCGATTCGAT
AGACTGCGTACT
AGCAGTCGCGAT
AGGACGCACTGT
AAGAGATGTCGA
ACAGCAGTGGTC
ACGTACTCAGTG
ACTCGCACAGGA
AGAGAGCAAGTG
AGCATATGAGAG
AGGCTACACGAC
AAGCTGCAGTCG
ACAGCTAGCTTG
ACCTGTCTCTCT
ACGTCTGTAGCA
AGAGCAAGAGCA
AGCCATACTGAC
AGGTGTGATCGC
AATCAGTCTCGT
ACGACGTCTTAG
ACGTGAGAGAAT
ACTGACAGCCAT
AGCGACTGTGCA
AGTACGCTCGAG
AATCGTGACTCG
ACGAGTGCTATC
ACTGATCCTAGT
AGAGTCCTGAGC
AGCGAGCTATCT
AGTACTGCAGGC
ACACACTATGGC
ACGATGCGACCA
ACGTTAGCACAC
ACTGTACGCGTA
AGATACACGCGC
AGCGCTGATGTG
ACACATGTCTAC
ACATGATCGTTC
ACGCAACTGCTA
ACTGTCGAAGCT
AGCGTAGGTCGT
ACACGAGCCACA
ACATGTCACGTG
ACGCGATACTGG
ACTACGTGTGGT
ACTGTGACTTCA
AGATCTCTGCAT
AGCTATCCACGA
AGTCCATAGCTG
ACACGGTGTCTA
ACATTCAGCGCA
ACTTGTAGCAGC
AGATGTTCTGCT
AGCTCCATACAG
ACGCTATCTGGA
ACTATTGTCACG
AGAACACGTCTC
AGTGAGAGAAGC
ATACTATTGCGC
ATGCACTGGCGA
CAGATACACTTC
CAGTGTCAGGAC
AGTGCGATGCGT
ATACTCACTCAG
ATCGCTCGAGGA
ATGCAGCTCAGT
ATTCTGTGAGCG
CACGGACTATAC
CAGATCGGATCG
CATACCAGTAGC
AGTGGATGCTCT
ATAGCTCCATAC
ATCGTACAACTC
CAACACGCACGA
CACGTCGATGGA
CAGCACTAAGCG
CATAGACGTTCG
AGTGTCACGGTG
ATAGGCGATCTC
ATCTACTACACG
ATGCGTAGTGCG
CATAGCGAGTTC
AGTGTTCGATCG
ATATCGCTACTG
ATCTCTGGCATA
ATGGATACGCTC
CAACTCATCGTA
CACTACTGTTGA
Table 1. Summary description of the sampling effort, the number of sequences collected, and the levels of bacterial
diversity discovered
No. of hands sampled
102 (from 27 men and 24 women)
Total no. of
sequences
Average length of
sequence reads,
bp (range)
Total no. of
classifiable
bacterial
sequences
Total no. of
phylotypes
across all hands
sampled
Average no.
of sequences
per hand (range)
Average no. of
phylotypes per
hand (range)
351,630
228 (200–267)
331,619
4,742
3,251 (2,410–5,838)
158 (46–401)
Phylotypes were determined at the 97% sequence similarity level.
the skin surface by at least an order of magnitude (8), confirming
that culture-based surveys of the skin surface, like surveys
conducted in many other microbial habitats (14), dramatically
underestimate the full extent of bacterial diversity. The average
phylotype richness observed on a single palm surface was also !3
times higher than the richness observed in a molecular survey of
forearm skin (6) and elbow skin (7). Although we would expect
the hand surface to have higher levels of diversity than other skin
surfaces because of the more frequent contact with potential
inocula from the environment, this discrepancy in observed
bacterial diversity is more likely a result of the depth of our
sampling, which allowed us to survey even those rare bacterial
taxa present on the skin surface. However, despite the depth of
our surveys, our diversity estimates still represent only the lower
bounds of phylotype richness on individual hands; the rarefaction curves for individual palm surfaces do not asymptote
[supporting information (SI) Fig. S1], indicating that the true
diversity is likely even higher. The total diversity of bacteria on
within the lower intestine (15–17), but this may be a function of
the depth of our sequencing. If we compare our results with those
obtained by Andersson et al. (18) where a similar pyrosequencing-based approach was used to survey human-associated bacterial communities, we find that skin bacterial communities
appear to be more diverse on average than those communities
found in throat, stomach, and fecal environments.
Although diversity on palm surfaces is high at both the
phylotype and phylum levels (sequences from !25 phyla were
detected), 3 phyla (Actinobacteria, Firmicutes, and Proteobacteria) accounted for 94% of the sequences (Fig. 1 and Table S1).
The most abundant genera (Proprionibacterium, 31.6% of all
sequences; Streptococcus, 17.2%; Staphylococcus, 8.3%; Corynebacterium, 4.3%; and Lactobacillus, 3.1%) were found on nearly
all palm surfaces sampled. These genera have previously been
found to be abundant in other molecular surveys of skin bacteria
(6, 19) and are considered to be common skin residents (5), yet
they still represented "65% of all of the identified sequences
Percentage of phylotypes shared ( , )
12
13
14
15
16
Average similarity between
hands from different
individuals
17
18
19
Average similarity
between hands from one
individual
A
B
C
~ 21 SHARED
SEQUENCES
OUT OF 158
0.30
~ 28 SHARED
SEQUENCES
OUT OF 158
0.32
0.34
0.36
D
0.38
UniFrac Similarity ( , )
Fig. 2. Average pairwise bacterial community similarity between left and
right hands from the same individual (circles) and between hands from
different individuals (squares) as measured by using the unweighted UniFrac
similarity index (bottom axis, open symbols) or the percentage of phylotypes
that are shared between pairs (top axis, filled symbols). Average pairwise
values and 95% confidence intervals are shown. For these analyses, 2,500
sequences were randomly selected per sample, and only those samples represented by "2,500 sequences were included (n # 51 and 5,100 pairwise
comparisons for intraindividual comparison and interindividual comparisons,
respectively).
0
Fig. 3. Differenti
dominant versus the
and time since last
unweighted UniFrac
degree of differenti
of the branch node
indicating that ea
communities.
A
, )
18
19
Female
Male
B
age similarity
hands from one
dividual
Dominant hand
Non-dominant hand
C
<2h
2-4h
>4 h
D
Female (< 2 h)
Male (< 2 h)
Male (2 - 4 h)
Female (2 - 4 h)
Female (>4 h)
Male (>4 h)
0.38
0.5
0.35
0.30
0.25
VARIATION WITHIN EACH GROUP
ilarity between
left and
(NOT DISTINGUISHABLE)
d between hands from
UniFrac Distance
the unweighted UniFrac
Fig. 3. Differentiation in hand-surface communities between sexes (A),
ercentage of phylotypes
dominant versus the nondominant hands (B), time since last hand washing (C),
bols). Average pairwise
and time since last hand washing for each sex (D) determined by using the
or these analyses, 2,500
unweighted UniFrac algorithm. The length of the branches corresponds to the
only those samples repdegree of differentiation between bacterial communities in each category. All
# 51 and 5,100 pairwise
of the branch nodes shown here were found to be significant (P ! 0.001),
rindividual comparisons,
indicating that each of these categories harbored distinct bacterial
communities.
relatively low abun-
0
, )
18
19
age similarity
hands from one
dividual
A
Female
Male
B
MALE AND FEMALE HANDS
ARE READILY DISTINGUISHED
Dominant hand
Non-dominant hand
C
<2h
2-4h
>4 h
D
Female (< 2 h)
Male (< 2 h)
Male (2 - 4 h)
Female (2 - 4 h)
Female (>4 h)
Male (>4 h)
0.38
ilarity between left and
d between hands from
the unweighted UniFrac
ercentage of phylotypes
bols). Average pairwise
or these analyses, 2,500
only those samples rep# 51 and 5,100 pairwise
rindividual comparisons,
relatively low abun-
0.35
0.30
0.25
UniFrac Distance
Fig. 3. Differentiation in hand-surface communities between sexes (A),
dominant versus the nondominant hands (B), time since last hand washing (C),
and time since last hand washing for each sex (D) determined by using the
unweighted UniFrac algorithm. The length of the branches corresponds to the
degree of differentiation between bacterial communities in each category. All
of the branch nodes shown here were found to be significant (P ! 0.001),
indicating that each of these categories harbored distinct bacterial
communities.
, )
18
19
age similarity
hands from one
dividual
A
Female
Male
B
Dominant hand
Non-dominant hand
C
DOMINANT/NON-DOMINANT
HANDS ARE READILY
DISTINGUISHED
<2h
2-4h
>4 h
D
Female (< 2 h)
Male (< 2 h)
Male (2 - 4 h)
Female (2 - 4 h)
Female (>4 h)
Male (>4 h)
0.38
ilarity between left and
d between hands from
the unweighted UniFrac
ercentage of phylotypes
bols). Average pairwise
or these analyses, 2,500
only those samples rep# 51 and 5,100 pairwise
rindividual comparisons,
relatively low abun-
0.35
0.30
0.25
UniFrac Distance
Fig. 3. Differentiation in hand-surface communities between sexes (A),
dominant versus the nondominant hands (B), time since last hand washing (C),
and time since last hand washing for each sex (D) determined by using the
unweighted UniFrac algorithm. The length of the branches corresponds to the
degree of differentiation between bacterial communities in each category. All
of the branch nodes shown here were found to be significant (P ! 0.001),
indicating that each of these categories harbored distinct bacterial
communities.
, )
18
19
age similarity
hands from one
dividual
A
Female
Male
B
Dominant hand
Non-dominant hand
C
<2h
2-4h
>4 h
D
HOW LONG IT’S BEEN SINCE
HANDWASHING IS
READILY DISTINGUISHED
Female (< 2 h)
Male (< 2 h)
Male (2 - 4 h)
Female (2 - 4 h)
Female (>4 h)
Male (>4 h)
0.38
ilarity between left and
d between hands from
the unweighted UniFrac
ercentage of phylotypes
bols). Average pairwise
or these analyses, 2,500
only those samples rep# 51 and 5,100 pairwise
rindividual comparisons,
relatively low abun-
0.35
0.30
0.25
UniFrac Distance
Fig. 3. Differentiation in hand-surface communities between sexes (A),
dominant versus the nondominant hands (B), time since last hand washing (C),
and time since last hand washing for each sex (D) determined by using the
unweighted UniFrac algorithm. The length of the branches corresponds to the
degree of differentiation between bacterial communities in each category. All
of the branch nodes shown here were found to be significant (P ! 0.001),
indicating that each of these categories harbored distinct bacterial
communities.
, )
18
19
age similarity
hands from one
dividual
A
Female
Male
B
Dominant hand
Non-dominant hand
C
<2h
2-4h
>4 h
D
Female (< 2 h)
Male (< 2 h)
Male (2 - 4 h)
Female (2 - 4 h)
THESE DISTINCTIONS
CAN BE COMBINED
Female (>4 h)
Male (>4 h)
0.38
ilarity between left and
d between hands from
the unweighted UniFrac
ercentage of phylotypes
bols). Average pairwise
or these analyses, 2,500
only those samples rep# 51 and 5,100 pairwise
rindividual comparisons,
relatively low abun-
0.35
0.30
0.25
UniFrac Distance
Fig. 3. Differentiation in hand-surface communities between sexes (A),
dominant versus the nondominant hands (B), time since last hand washing (C),
and time since last hand washing for each sex (D) determined by using the
unweighted UniFrac algorithm. The length of the branches corresponds to the
degree of differentiation between bacterial communities in each category. All
of the branch nodes shown here were found to be significant (P ! 0.001),
indicating that each of these categories harbored distinct bacterial
communities.
Fig. S3. Differentiation in hand-surface communities with time since last hand washing for each sex as determined from the smaller-scale study of 8 individuals.
Community differentiation was measured by using the unweighted UniFrac algorithm; the length of the branches corresponds to the degree of differentiation
between bacterial communities in each category. All of the branch nodes shown here were found to be significant (P ! 0.001), indicating that each of these 8
categories harbored distinct bacterial communities.
MALE AND FEMALE HANDS
ARE MOST SIMILAR
RIGHT AFTER WASHING
Fig. S3. Differentiation in hand-surface communities with time since last hand washing for each sex as determined from the smaller-scale study of 8 individuals.
Community differentiation was measured by using the unweighted UniFrac algorithm; the length of the branches corresponds to the degree of differentiation
between bacterial communities in each category. All of the branch nodes shown here were found to be significant (P ! 0.001), indicating that each of these 8
categories harbored distinct bacterial communities.
... THEN QUICKLY BECOME
MORE DISTINCT
Fig. S3. Differentiation in hand-surface communities with time since last hand washing for each sex as determined from the smaller-scale study of 8 individuals.
Community differentiation was measured by using the unweighted UniFrac algorithm; the length of the branches corresponds to the degree of differentiation
between bacterial communities in each category. All of the branch nodes shown here were found to be significant (P ! 0.001), indicating that each of these 8
categories harbored distinct bacterial communities.
A
Average phylodiversity
per hand
12
PHYLODIVERSITY =
SUM LENGTH OF ALL
BRANCHES IN THE TREE
10
8
6
Male
Female
4
2
Average number of unique
phylotypes per hand
B
175
150
125
100
75
50
25
500
1000
1500
2000
2500
Number of sequences (standardized )
Fig. 4. Rarefaction curves showing differences in bacterial diversity on palm
surfaces from men and women. (A) Phylogenetic diversity estimated by measuring the average total branch length per sample after a specified number of
individual sequences have been observed (36). (B) Diversity estimated by
determining the average number of unique phylotypes per hand. For these
analyses, we randomly selected 2,400 sequences per hand sample, and thus
the average number of phylotypes per hand is lower than for the full dataset
(Table 1). Bars indicate 95% confidence intervals.
abundant (or did not change appreciably
time since last hand washing were less abu
women (Fig. 1). Likewise, even if we separ
both sex and by hand-washing categor
significant differences between the sexes
To further resolve the effects of sex and
palm bacterial communities, we conducte
men and 4 women to explicitly examine th
of skin bacterial communities after hand w
more controlled because we did not re
estimate of the time since last hand wash
palms of each of these 8 individuals every
after hand washing. Also, unlike the large
1 swab to sample the communities on bo
hands
of each
individual,
and the volu
FEMALE
HANDS
HARBOR
during
a
normal
work
day,
not immedi
MORE MICROBIAL DIVERSITY
examination where student anxiety may h
bacterial communities. However, we found
in the 2 different studies. Specifically, we
bacterial community composition with
washing were significant and nearly identi
above for the larger study (Table S2 an
confirmed that men and women harbor d
munities, even when controlling for hand h
differences between the sexes become mo
since hand washing (Fig. S3). Likewise
women do harbor higher levels of bacter
hands than men (Fig. S4).
Together these results demonstrate the
sequencing technologies to survey microb
unprecedented level of detail. There appe
phylotypes present on the skin of the adult
genomes of representatives of these organ
itized for sequencing to make sense of
studies. However, the noncore phylotype
‘‘long tail’’ effect—most phylotypes are
exhaustive sampling is not a reasonable go
significant heterogeneity in community
left and right hands from the same ind
careful sampling strategies will be required
for the International Human Microbiome
the relative numbers of core and noncore
Fig. S1. (A) Rarefaction curves from three individual hand samples, selected to be representative of individual palms with low, average, and high levels of
bacterial diversity. (B and C) Rarefaction curves for samples grouped into categories based on time since last hand washing (B) and the dominant versus
nondominant hands (C). Curves were estimated by randomly selecting 2,400 sequences per hand sample so the average number of phylotypes per hand is lower
than that estimated for the full dataset. The number of individual hand samples included in each category is indicated in the legend. Confidence intervals are
shown at the 95% level.
THE LEVEL OF DIVERSITY
BETWEEN HANDS IS
USUALLY SIMILAR
Fig. S1. (A) Rarefaction curves from three individual hand samples, selected to be representative of individual palms with low, average, and high levels of
bacterial diversity. (B and C) Rarefaction curves for samples grouped into categories based on time since last hand washing (B) and the dominant versus
nondominant hands (C). Curves were estimated by randomly selecting 2,400 sequences per hand sample so the average number of phylotypes per hand is lower
than that estimated for the full dataset. The number of individual hand samples included in each category is indicated in the legend. Confidence intervals are
shown at the 95% level.
THE LEVEL OF DIVERSITY
PER HAND
DOESN’T CHANGE
SIGNIFICANTLY WITH
WASHING
Fig. S1. (A) Rarefaction curves from three individual hand samples, selected to be representative of individual palms with low, average, and high levels of
bacterial diversity. (B and C) Rarefaction curves for samples grouped into categories based on time since last hand washing (B) and the dominant versus
nondominant hands (C). Curves were estimated by randomly selecting 2,400 sequences per hand sample so the average number of phylotypes per hand is lower
than that estimated for the full dataset. The number of individual hand samples included in each category is indicated in the legend. Confidence intervals are
shown at the 95% level.
THERE IS NO DIFFERENCE
IN DIVERSITY BETWEEN
DOMINANT AND NON-DOMINANT
HANDS
Fig. S1. (A) Rarefaction curves from three individual hand samples, selected to be representative of individual palms with low, average, and high levels of
bacterial diversity. (B and C) Rarefaction curves for samples grouped into categories based on time since last hand washing (B) and the dominant versus
nondominant hands (C). Curves were estimated by randomly selecting 2,400 sequences per hand sample so the average number of phylotypes per hand is lower
than that estimated for the full dataset. The number of individual hand samples included in each category is indicated in the legend. Confidence intervals are
shown at the 95% level.
Fig. S4. Rarefaction curves for samples grouped into categories based on sex (a) and time since last hand washing (b) from the smaller-scale study. Notice that
the number of sequences collected is far less than the number collected for the main study. Confidence intervals are shown at the 95% level. Results are from
the 4 men and 4 women sampled immediately after an initial hand washing (0 h) and every 2 h thereafter for a 6-h period.
NOT ONLY ARE FEMALES
MORE DIVERSE THAN
MALES PER HAND,
THEY’RE MORE DIVERSE
PER INDIVIDUAL
Fig. S4. Rarefaction curves for samples grouped into categories based on sex (a) and time since last hand washing (b) from the smaller-scale study. Notice that
the number of sequences collected is far less than the number collected for the main study. Confidence intervals are shown at the 95% level. Results are from
the 4 men and 4 women sampled immediately after an initial hand washing (0 h) and every 2 h thereafter for a 6-h period.
DIVERSITY IS NOT HIGHER
PER HAND, BUT IS MORE
DIVERSE PER INDIVIDUAL
RIGHT AFTER WASHING
HANDS ARE MOST DIFFERENT
FROM EACH OTHER RIGHT
AFTER WASHING, THEN
QUICKLY BECOME MORE ALIKE
Fig. S4. Rarefaction curves for samples grouped into categories based on sex (a) and time since last hand washing (b) from the smaller-scale study. Notice that
the number of sequences collected is far less than the number collected for the main study. Confidence intervals are shown at the 95% level. Results are from
the 4 men and 4 women sampled immediately after an initial hand washing (0 h) and every 2 h thereafter for a 6-h period.
WHO’S ON YOUR HANDS?
KEEP IN MIND THAT THESE RESULTS WILL REFLECT
PRIMER SPECIFICITY AND PCR BIAS
Fig. S2. Relative abundances of the most abundant bacterial g0 0roups on the hand surfaces sampled as part of the smaller scale study, with the hand samples
divided into categories of sex (A) and time since last hand washing (B). Four men and 4 women were sampled every 2 h for a 6-h period after hands were
thoroughly washed. Error bars are one standard error of the mean. For the number of sequences and number of samples included in each category and the full
taxonomic description of the hand-surface bacterial communities see Table S2. Superscripts on the taxon name indicate the phylum or subphylum: 1,
Actinobacteria; 2, Firmicutes; 3, Betaproteobacteria; 4, Gammaproteobacteria; 5, Alphaproteobacteria.
Table S2. Relative abundances of the bacterial groups from the palm surfaces sampled in the smaller-scale study designed to
specifically examine the influence of hand washing on bacterial community composition
Sex
Female
Male
0 hours
Time since hand washing
2 hours
4 hours
6 hours
16
16
8
8
8
8
5628
7547
2971
3116
3608
3480
Acidobacteria
0.11 (0.05)
0.02 (0.02)
0.00 (0.00)
0.09 (0.09)
0.10 (0.05)
0.06 (0.04)
Actinobacteria
Actinomycineae
0.36 (0.14)
0.10 (0.06)
0.35 (0.26)
0.17 (0.09)
0.28 (0.15)
0.11 (0.06)
Corynebacterium
2.67 (0.50)
3.58 (0.78)
5.91 (1.25)
2.43 (0.48)
2.40 (0.37)
1.75 (0.63)
Frankineae
0.35 (0.11)
0.07 (0.04)
0.18 (0.11)
0.40 (0.20)
0.14 (0.07)
0.12 (0.06)
No. of individual swabs collected
No. of sequences
Bacteroidetes
Alphaproteobacteria
Betaproteobacteria
Intrasporangiaceae
0.19 (0.13)
0.24 (0.14)
0.31 (0.28)
0.31 (0.26)
0.15 (0.07)
0.09 (0.07)
Proprionibacterium
57.85 (5.81)
65.56 (5.08)
38.34 (5.18)
68.06 (6.79)
70.64 (5.65)
69.79 (7.37)
Other
2.52 (0.39)
1.68 (0.30)
2.89 (0.47)
1.76 (0.45)
1.62 (0.44)
2.12 (0.63)
Capnocytophaga
0.19 (0.07)
0.05 (0.03)
0.11 (0.07)
0.09 (0.06)
0.16 (0.12)
0.13 (0.05)
Chryseobacterium
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
Hymenobacter
0.33 (0.11)
0.10 (0.05)
0.18 (0.12)
0.26 (0.09)
0.11 (0.09)
0.31 (0.20)
Porphyromonas
0.41 (0.16)
0.25 (0.17)
0.17 (0.08)
0.39 (0.27)
0.22 (0.16)
0.56 (0.34)
Prevotella
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
Saprospirales
0.41 (0.23)
0.03 (0.02)
0.51 (0.46)
0.18 (0.08)
0.02 (0.02)
0.15 (0.08)
Other
1.09 (0.29)
0.84 (0.28)
2.40 (0.34)
0.55 (0.22)
0.59 (0.34)
0.31 (0.14)
Acetobacterales
0.10 (0.05)
0.14 (0.06)
0.24 (0.13)
0.07 (0.07)
0.06 (0.04)
0.11 (0.06)
Bradyrhizobiales
1.56 (0.49)
0.91 (0.26)
2.45 (0.68)
0.68 (0.25)
0.99 (0.62)
0.81 (0.42)
Caulobacterales
0.21 (0.10)
0.13 (0.06)
0.47 (0.18)
0.06 (0.04)
0.03 (0.03)
0.13 (0.06)
Rhizobiaceae
0.04 (0.02)
0.07 (0.04)
0.20 (0.08)
0.00 (0.00)
0.00 (0.00)
0.02 (0.02)
Rhodobacterales
0.23 (0.10)
0.16 (0.07)
0.08 (0.04)
0.32 (0.19)
0.10 (0.07)
0.29 (0.10)
Sphingomonadales
0.57 (0.19)
0.50 (0.14)
0.99 (0.32)
0.48 (0.17)
0.31 (0.14)
0.38 (0.20)
Other
0.27 (0.08)
0.03 (0.02)
0.13 (0.10)
0.29 (0.12)
0.08 (0.06)
0.10 (0.08)
Burkholderiales
3.41 (1.22)
2.00 (0.75)
8.07 (1.82)
1.15 (0.45)
0.80 (0.20)
0.79 (0.23)
Neisseriales
0.95 (0.28)
0.51 (0.15)
0.63 (0.26)
0.79 (0.44)
0.93 (0.36)
0.58 (0.26)
Other
0.25 (0.14)
0.20 (0.08)
0.64 (0.25)
0.15 (0.10)
0.02 (0.02)
0.08 (0.08)
Myxococcales
0.12 (0.08)
0.00 (0.00)
0.19 (0.16)
0.04 (0.04)
0.00 (0.00)
0.00 (0.00)
Other
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
Epsilonproteobacteria
Other
0.37 (0.30)
0.00 (0.00)
0.10 (0.10)
0.05 (0.05)
0.60 (0.60)
0.00 (0.00)
Gammaproteobacteria
Enterobacteriales
1.15 (0.55)
0.26 (0.19)
0.41 (0.29)
1.44 (0.98)
0.64 (0.59)
0.32 (0.24)
Moraxellaceae
2.67 (1.00)
3.22 (0.97)
8.24 (1.59)
1.10 (0.33)
1.42 (0.42)
1.02 (0.48)
Pasteurellaceae
1.87 (0.48)
1.37 (0.45)
1.37 (0.48)
1.20 (0.62)
1.60 (0.70)
2.30 (0.82)
Pseudomonadaceae
0.94 (0.34)
1.85 (0.78)
2.38 (0.58)
2.00 (1.54)
0.89 (0.29)
0.29 (0.17)
Xanthomonadales
0.10 (0.06)
0.09 (0.07)
0.28 (0.15)
0.10 (0.07)
0.00 (0.00)
0.00 (0.00)
Other
0.19 (0.11)
0.07 (0.04)
0.07 (0.05)
0.32 (0.22)
0.09 (0.07)
0.04 (0.04)
Chloroflexi
0.01 (0.01)
0.03 (0.02)
0.03 (0.03)
0.00 (0.00)
0.03 (0.03)
0.03 (0.03)
Chloroplasts
2.01 (0.49)
2.96 (1.06)
1.11 (0.56)
2.89 (1.35)
3.59 (1.42)
2.36 (1.17)
Cyanobacteria
0.11 (0.08)
0.00 (0.00)
0.16 (0.16)
0.00 (0.00)
0.06 (0.06)
0.00 (0.00)
Firmicutes
Acidaminococcaceae
0.71 (0.15)
0.26 (0.10)
0.66 (0.22)
0.47 (0.20)
0.50 (0.20)
0.31 (0.18)
Aerococcaceae
0.40 (0.14)
0.30 (0.08)
0.52 (0.19)
0.19 (0.13)
0.27 (0.11)
0.43 (0.19)
Brochothrix
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
Lactobacillaceae
2.69 (0.93)
0.88 (0.26)
2.03 (0.70)
2.62 (1.80)
0.90 (0.51)
1.59 (0.50)
Peptostreptococcaceae
1.21 (0.35)
0.55 (0.21)
1.89 (0.56)
0.52 (0.33)
0.44 (0.18)
0.66 (0.34)
Staphylococcaceae
3.49 (0.61)
2.84 (0.73)
6.27 (1.25)
2.47 (0.38)
2.15 (0.37)
1.78 (0.46)
Streptococcaceae
5.27 (0.78)
6.87 (1.57)
6.28 (1.83)
4.22 (1.46)
5.18 (0.98)
8.60 (2.36)
Other
1.16 (0.20)
0.63 (0.15)
1.11 (0.21)
0.89 (0.25)
0.99 (0.21)
0.58 (0.36)
Fusobacteria
0.40 (0.14)
0.40 (0.11)
0.67 (0.18)
0.36 (0.23)
0.22 (0.07)
0.36 (0.20)
Gemmatimonadetes
0.03 (0.03)
0.00 (0.00)
0.05 (0.05)
0.00 (0.00)
0.00 (0.00)
0.00 (0.00)
Thermi
0.85 (0.38)
0.08 (0.04)
0.84 (0.73)
0.29 (0.18)
0.23 (0.20)
0.50 (0.26)
TM7
0.07 (0.03)
0.06 (0.03)
0.07 (0.05)
0.03 (0.03)
0.10 (0.05)
0.05 (0.04)
Other
0.11 (0.08)
0.11 (0.07)
0.00 (0.00)
0.13 (0.13)
0.32 (0.14)
0.00 (0.00)
Deltaproteobacteria
All of the abundances are reported as percentages of the sequences within each category that match the taxonomic group with one standard error of the mean
indicated in parentheses. For this table, we used the Hugenholtz classification scheme against the Greengenes database (10), and the sequences were classified to
the level of taxonomic resolution deemed to be most appropriate. For this study, we swabbed both left and right hands together from 4 men and 4 women
immediately after hand washing (time 0) and every 2 h over a 6-h period.
WHAT DOES ALL THIS MEAN?
PROPRIONOBACTERIA, AND THE
OTHER USUAL SUSPECTS,
PREDOMINATE HUMAN HAND
MICROFLORA
WHAT DOES ALL THIS MEAN?
PROPRIONOBACTERIA, AND THE
OTHER USUAL SUSPECTS,
PREDOMINATE HUMAN HAND
MICROFLORA
WITHIN THIS CONTEXT, THERE ARE
LARGE VARIATIONS BETWEEN
SEXES, HAND WASHINGS,
INDIVIDUALS, AND EVEN HANDS OF
THE SAME INDIVIDUAL
Fig. S4. Rarefaction curves for samples grouped into categories based on sex (a) and time since last hand washing (b) from the smaller-scale study. Notice that
the number of sequences collected is far less than the number collected for the main study. Confidence intervals are shown at the 95% level. Results are from
the 4 men and 4 women sampled immediately after an initial hand washing (0 h) and every 2 h thereafter for a 6-h period.
Other Supporting Information Files
Table S1 (PDF)
Table S2 (PDF)
Table S3 (PDF)
Fierer et al. www.pnas.org/cgi/content/short/0807920105
4 of 4
WHAT DOES ALL THIS MEAN?
PROPRIONOBACTERIA, AND THE
OTHER USUAL SUSPECTS,
PREDOMINATE HUMAN HAND
MICROFLORA
WITHIN THIS CONTEXT, THERE ARE
LARGE VARIATIONS BETWEEN
SEXES, HAND WASHINGS,
INDIVIDUALS, AND EVEN HANDS OF
THE SAME INDIVIDUAL
DEFINING “NORMAL” MICROFLORA
AND DISTINGUISHING THIS FROM
“ABNORMAL” IS A DIFFICULT
QUANTITATIVE/STATISTICAL
PROBLEM