Submitted on Wednesday, September 25, 2019 - 09:14
Submitted by anonymous user: 134.50.65.234
Submitted values are:

Title of Dataset: The microbial associates of precipitation over four
seasons, at three locations in the United States
Description:
These data were used in a paper accepted in Ecological Monographs on 17
September 2019, entitled "Spatiotemporal patterns of microbial composition
and diversity
in precipitation", hereafter Aho et al. (in press).  Microbes in the
atmosphere have broad ecological impacts, including the potential to trigger
precipitation through species and strains that act as ice nucleation
particles. To characterize spatiotemporal trends of microbial assemblages in
precipitation we sequenced 16S (bacterial) and 18S (fungal) rRNA gene
amplicon libraries collected from 72 precipitation events in three US states
(Idaho, Louisiana, and Virginia) over four seasons.

The datasets provided here include:
1) Read totals for fungal and bacterial operational taxonomic units (OTUs),
2) Taxonomic assignments at levels of domain through menus for individual
OTUs,
3) Relevant metadata.

Methodological Details.  Text follows Aho et al. (in press):
  From May 2013 to July 2015 a total of 72 distinct precipitation events were
sampled at locations associated with three participating U.S. universities:
20 m Above Ground Level (mAGL) on the roof of the Life Sciences Building at
Louisiana State University (LSU) in Baton Rouge, LA, USA (30° 25’ 12” N,
91° 10' 48" W; 11 m Height Above Ellipsoid (HAE)), 20 mAGL on the  roof of
the Gale Life Sciences Building at Idaho State University (ISU) in Pocatello,
ID, USA (42° 52’ 3.85” N, 112° 25’ 47.23” W; 1440 m HAE), 10 mAGL
on the roof of Latham Hall at Virginia Tech (VT) in Blacksburg, VA (37°
13’ 29” N, 80° 25’ 22” W; 627 m HAE) and at three sites at VT’s
Kentland Farm: site A (37° 12’ 9” N, 80° 33’ 51” W; 515 m HAE),
site B (37° 11’ 47” N, 80° 34’ 41” W; 531 m HAE) and site C (37°
12’ 8” N, 80° 35’ 32” W; 594 m HAE) (see Table 1).  Sample dates
were placed into the following seasonal categories: spring (March 29-June
12), summer (June 25-Sept 6), winter (Dec 22-March 16) and autumn (Oct 12 -
Dec 19) .  Sixty-six of the 72 precipitation events were rainstorms and the
remaining six events were ice/snow storms.  Ice/snow storms occurred at all
three sites.

Bacterial 16S and fungal 18S rRNA gene amplicon libraries were sequenced from
precipitation samples collected during all four seasons at LSU, and three
seasons at ISU (spring, summer and winter), and VT (summer, winter and
autumn) (Aho et al. in press; Table 1).  Relatively rare heavy-precipitation
events contributed to the smaller number of samples at ISU.   The 30
precipitation events sampled for 18S rRNA genes constituted a subset of the
71 events sampled for 16S except for one case (LSU-spring) in which a
precipitation event was sampled for 18S but not 16S rRNA genes (Aho et al. in
press; Table 1).  The subset of 18S analyses was due to the depletion of
sample materials from the 16S rRNA gene amplification procedures, which
generally preceded 18S rRNA gene sequencing. In 29 precipitation events both
18S and 16S rRNA genes were successfully amplified and sequenced.

For each sampling event, precipitation was collected in ~120 L galvanized
collection cans that were lined with clean, sterile 94 x 122 cm polypropylene
bags (Fisher Scientific, Pittsburgh, PA).  A minimum volume of 3L was
amalgamated from collection cans and sample collection was conducted using a
total of eleven cans, ten for samples and one as a procedural control (see
Aho et al. in press, Appendix S1, Fig S1).  Immediately following each
precipitation event, collected material was transferred to sterile 9L carboys
(Aho et al. in press, Appendix S1, Fig S2) or left in the collection bags and
stored at 4 °C preceding concentrating by filtration, which was always
completed within 48 h of precipitation collection.  Precipitation was
filtered (1-3 L) onto 0.2 µm, 47 mm Supor PES membrane filters or Millipore
polycarbonate filters.  Filters were placed into sterile 15 mL or 50 mL
falcon tubes and frozen preceding further processing.  All samples from
Louisiana and Virginia were shipped overnight on dry ice to ISU for
consolidation with ISU samples.

Sampling procedures were designed to ensure that 16S and 18S rRNA gene
sequences were not obtained from contaminated sampling equipment, sampling
processes, or downstream molecular procedures.  Procedural controls were
created by lining one galvanized steel can with a sterile polypropylene bag
(just as was done for precipitation sample collection) and covering the can
with an ethanol-cleaned lid (Aho et al. in press,  Appendix S1, Fig S1).  One
L of autoclaved nanopure water was poured into the control bag and this water
was filtered in the same manner as the precipitation samples.  All filters
containing procedural controls were put through the same DNA extraction and
gene amplification procedures to assess spurious amplification of 16S and 18S
rRNA genes due to procedural and/or reagent contamination.

DNA extraction from precipitation samples
Microbial wet lab procedures, including choices of primers, followed
guidelines of the Earth Microbiome Project (Thompson et al. 2017).  In
preparation for DNA extraction, filters were cut using ethanol cleaned and
flamed scissors.  All filter cutting was performed in a laminar flow hood and
filters were only handled with sterile tweezers.  Each filter was cut into
the smallest possible pieces (< 5mm in length) just above a bead beating tube
from the FastDNA SPIN Kit for Soil (MP Biomedicals, USA).  Once in the bead
beating tube, the manufacturer’s protocol for the FastDNA SPIN Kit for Soil
was followed, starting with the addition of MT and phosphate buffer
solutions.  The manufacturer’s protocol was followed with the exception of
the following steps: in step 4, a mini bead beater was used to homogenize
extracts for 70 seconds, step 5 was performed for 8 minutes, a 15 mL tube was
always used for step 7, and 100 µL of DNase/pyrogen-free water was used to
elute DNA in step 16.  The resulting extracts were further purified with the
aid of a silica spin filter, following steps 14-22 of the protocol for the
MoBio Power Soil Kit (MoBio Laboratories, Carlsbad, CA).  In these steps, 160
µL of solution C4 was added to the DNA extraction.  The manufacturer’s
protocol was followed for other steps, with the exception of adding 25 µL of
solution C6 and incubating for 1 minute at room temperature prior to
centrifugation.  All DNA extracts were stored at -20 °C until PCR
amplification of the 16S and 18S rRNA genes was attempted.

Bacterial 16S rRNA gene fragment amplification
The V4 regions of bacterial 16S rRNA gene fragments were PCR-amplified in
triplicate from each DNA extract for sequencing.  PCR amplification was
carried out using barcoded primers following the methods of Caporaso et al.
(2012).  The 25 µL PCR mix contained the following: 1X 5-PRIME HotMaster
Mix, 0.2 µM 515F, 0.2 µM 806R and nuclease free water (13 µL).
Thermal-cycling was carried out using 1-6 µL of DNA template in an Eppendorf
PRO S Master Cycler under the following conditions: initial denaturation at
94 °C for 3 minutes, 32 cycles of denaturation at 94 °C for 1 minute,
annealing at 50 °C for 1 minute and extension at 72 °C for 1:45, followed
by a final extension at 72 °C for 10 minutes.  We resolved
electrophoretically separated PCR products on a 1% TAE gel stained with
ethidium bromide to confirm successful amplifications in each reaction.
Triplicate reactions were pooled before purification and concentration using
the Qiagen MinElute PCR Cleanup kit (Qiagen, Valencia, California, USA).  A
35% guanidine-HCl wash step was performed to ensure that large primer-dimers
were removed from the samples.  Amplicons were stored at -20 °C preceding
sequencing at Idaho State University’s Molecular Research Core Facility
(Pocatello, ID, USA) on the Illumina MiSeq Platform using the V2 500 bp kit.

Fungal 18S rRNA gene fragment amplification
Fungal 18S rRNA gene fragments from the V9 region were PCR-amplified in
triplicate from each DNA extract.  Steps for 16S amplification were also
followed for 18S amplification with two exceptions.   First, the 25 µL PCR
mix contained 1X 5-PRIME HotMaster Mix, 0.2 µM 1391F, 0.2 µM EukBr and
nuclease free water (13 µL).  Second, thermal cycling used the following
conditions: initial denaturation at 94 °C for 3 minutes, 32 cycles of
denaturation at 94 °C for 0:45, annealing at 57 °C for 1 minute and
extension at 72 °C for 1:30, followed by a final extension at 72 °C for 10
minutes.

Sequence data analysis
Sequence data were analyzed using the mothur (ver. 1.42.3) software package
(Schloss et al. 2009). Contigs were assembled and parsed on the basis of
unique barcodes attached to the 806R primer (Caporaso et al. 2012).
Sequences that did not contain exact matches to the primer and barcodes
utilized in the PCR amplification were discarded. Sequences were filtered for
quality with a 50-base sliding window.  Sequences not obtaining an average
quality score of 25, containing ambiguous bases, homopolymers (> 7 bases),
having lengths > 259 bases, or deemed chimeric using the UCHIME algorithm
(Edgar et al. 2011) were eliminated.
All 16S rRNA gene sequences were aligned to the SILVA bacterial 16S rRNA gene
reference (release 132) (Pruesse et al. 2007) truncated to contain the
amplified V4 hypervariable region.  Sequences were taxonomically classified
using the classify.seqs command with arguments set to default values, with
the exception of the minimum consensus confidence value being set at 80%.
All 16S rRNA gene sequences that were classified as eukaryotic, archaea,
mitochondria, or chloroplast were removed from analyzed datasets.  Further,
pseudoreplicate 16S rRNA gene samples with fewer than 40,000 total reads,
following removal of non-informative OTUs, were eliminated from downstream
analyses, and thus were not averaged with other pseudoreplicates to obtain a
single independent 16S observation per storm (see Appendix S1: Table S2).
The 16S rRNA gene sequence library size was normalized by randomly sampling a
total of 44,621 sequences from each library, which was the number of
sequences in the smallest library.

The sequences of 18S rRNA gene amplicons for each sample were aligned in
mothur to the SILVA 18S rRNA gene reference (release 132), truncated to
contain the V9 hypervariable region, with minimum consensus confidence set at
80%.  Non-fungal sequences were eliminated from the 18S dataset.  Analogous
to methods for bacteria, pseudoreplicate fungal samples with fewer than
20,000 reads were eliminated. Average read totals for events were
approximately 40% lower for 18S than for 16S rRNA gene sequence libraries.
The 18S sequence library size was normalized by randomly sampling a total of
28,586 sequences from each library; the number of sequences in the smallest
library.

Both 16S and 18S rRNA gene sequences were clustered into operational
taxonomic units (OTUs) based on a sequence dissimilarity of ≤ 0.03, using
the Needleman-Wunsch algorithm (Kunin et al. 2010, Smith and Peay 2014).
Following the quality control procedures described above, the 16S rRNA gene
dataset contained 34,107 OTUs describing 71 independent rain event samples,
and the 18S rRNA gene dataset contained 16,597 OTUs describing 30 independent
rain event samples.  Both 16S and 18S rRNA gene sequence libraries were
deposited into MG-RAST metagenomics RAST (Rapid Annotation using Subsystem
Technology) server (https://metagenomics.anl.gov).  Accession numbers for
sequences are given in (Aho et al in press, Appendix S1, § S1.6.1).

Collection of Meteorological Data
Storm structure classifications were based on cloud top data retrieved from
The National Weather Service (NWS) archive of Geostationary Operational
Environmental Satellite-East (GOES-East) infrared satellite imagery, and Next
Generation Radar (NEXRAD). Level II and III radar reflectivities were
provided by the National Climatic Data Center’s (NCDC) website
(https://www.ncdc.noaa.gov/nexradinv/). Troposphere temperature profiles were
retrieved from The NWS radiosonde data archive of stations located in Lake
Charles, Louisiana (KLC) Slidell Muni, Louisiana (KLI), Pocatello, Idaho
(KSFX), and Roanoke Virginia (KFCX).

Seventy-two-hour backward trajectories of air masses present over Baton
Rouge, LA, Blacksburg, VA, and Pocatello, ID at times of collection were
calculated using the NOAA Air Resources Laboratory Hybrid Single-Particle
Lagrangian Integrated Trajectory (HYSPLIT) model.  Altitudes of 100, 1000,
5000, 6000, 8000 and 9000 mAGL were chosen for trajectory examination.   The
9000 m AGL upper altitudinal bound was chosen due to its proximity to the
homogeneous freezing layer (~ -38°C), where ice nucleation by heterogeneous
mechanisms is no longer necessary for the freezing of water molecules.

Classification of convective vs. stratiform precipitation was based on the
methods of Biggerstaff and Listemaa (2000), and Anagnostou (2004), in
addition to the inspection of tropospheric stability indices from National
Weather Service soundings (NWS 2018) https://www.weather.gov/lmk/indices),
and infrared and visible satellite imagery obtained from GOES-East (NOAA
GOES-East 2018).  Stratiform precipitation is defined specifically as
precipitation collected from stratus and nimbostratus-like cloud systems
independent of trailing stratiform regions from convective storm formations
(Biggerstaff and Listemaa 2000). Precipitation events were also assigned to
six morphological classes: Air mass, Mesoscale Convective System (MCS),
Multicell, Nimbostratus, and Squall.

Precipitation event trajectories were classified into five storm origin
categories based on a conventional scheme for North American air masses (AMS
2018). Seventy-two-hour trajectories, which originated over marine surfaces,
were classified as maritime (m), whereas those that originated over
terrestrial surfaces were classified as continental (c). Likewise, those
events that originated at latitudes above the subtropical regions of the
Northern Hemisphere (> 23.5°N) were classified as Polar (P), and those
originating in the subtropical and tropical regions of the Northern
Hemisphere (< 23.5°N) were classified as tropical (T). Thus, six possible
air mass combination classes were considered: Pacific Maritime Polar (mP1),
Atlantic Maritime Polar (mP2), Pacific Maritime Tropical (mT1), Atlantic
Maritime Tropical (mT2), Continental Polar (cP), and Continental Tropical
(cT).

References:

Aho, K., Weber, C. F., Christner, B. C., Vinatzer, B. A., Morris, C. E.,
Joyce, R., Failor, K., Werth, J. T., Bayless-Edwards, A. L. H., Schmale III,
D. G.  In press 9/17/2019.  Spatiotemporal patterns of microbial composition
and diversity in precipitation. Ecological Monographs ECM18-0180.R1

AMS (American Meteorological Society).  2018. Airmass Classification.
http://glossary.ametsoc.org/wiki/Airmass_classification (Accessed
12/18/2018).

Anagnostou, E. N.  2004. A convective/stratiform precipitation classification
algorithm for volume scanning weather radar observations. Meteorological
Applications 11: 291-300.

Biggerstaff, M. I., and S. A. Listemaa. 2000. An improved scheme for
convective/stratiform echo classification using radar reflectivity. Journal
of Applied Meteorology 39: 2129-2150.

Caporaso, J. G., C. L. Lauber, W. A. Walters, D. Berg-Lyons, J. Huntley, N.
Fierer, S. M. Owens, J. Betley, L. Fraser, M. Bauer, N. Gormley, J. A.
Gilbert, G. Smith, and R. Knight. 2012. Ultra-high-throughput microbial
community analysis on the Illumina HiSeq and MiSeq platforms. The ISME
journal 6: 1621-1624.
Edgar, R. C., B. J. Haas, J. C. Clemente, C. Quince, and R. Knight. 2011.
UCHIME improves sensitivity and speed of chimera detection. Bioinformatics
27: 2194-2200.

Kunin, V., A. Engelbrektson, H. Ochman, and P. Hugenholtz. 2010. Wrinkles in
the rare biosphere: pyrosequencing errors can lead to artificial inflation of
diversity estimates. Environmental Microbiology 12: 118-123.

NOAA GOES-East. 2018.  https://www.star.nesdis.noaa.gov/GOES/index.php
(Accessed 7/3/2018)

NWS. 2018. https://www.weather.gov/lmk/indices (Accessed 7/3/2018)

Pruesse, E., C. Quast, K. Knittel, B. M. Fuchs, W. Ludwig, J. Peplies, and F.
O. Glöckner.  2007. SILVA: a comprehensive online resource for quality
checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic
Acids Research 35: 7188-7196.

Smith, D. P., and K. G. Peay. 2014. Sequence depth, not PCR replication,
improves ecological inference from next generation DNA sequencing. PLoS One
9: e90234.

Thompson, L. R., J. G. Sanders, D. McDonald, A. Amir, J. Ladau, K. J. Locey,
R. J. Prill, A. Tripathi, S. M. Gibbons, G. Ackermann, J. A. Navas-Molina, S.
Janssen, E. Kopylova, Y. Vázquez-Baeza1, A. González, J. T. Morton, S.
Mirarab, Z. Z. Xu, L. Jiang, M. F. Haroon, J. Kanbar, Q. Zhu, S. J. Song, T.
Kosciolek, N. A. Bokulich, J. Lefler, C. J. Brislawn, G. Humphrey, S. M.
Owens, J. Hampton-Marcell, D. Berg-Lyons, V. McKenzie, N. Fierer, J. A.
Fuhrman, A. Clauset, R. L. Stevens, A. Shade, K. S. Pollard, K. D. Goodwin,
J. K. Jansson, J. A. Gilbert, R. Knight, and the Earth Microbiome Project
Consortium. 2017. A communal catalogue reveals Earth’s multiscale microbial
diversity. Nature, 551:  457–463

Authors:
K. Aho, C. F. Weber, J. T. Werth, A. L. H. Bayless-Edwards, Biological
Sciences, Idaho State University, Pocatello, ID 83209-8007
R. Joyce, B. C. Christner, Department of Microbiology and Cell Science,
Biodiversity Institute, University of Florida, Gainesville, FL 32611
K. Failor, B. A. Vinatzer, D. G. Schmale III, School of Plant and
Environmental Sciences, Virginia Tech, Blacksburg, VA 24061-0331.
C. E. Morris, INRA, Plant Pathology Research Unit 407, Montfavet, France.
Contact Person: Ken Aho
Contact Email: ahoken@isu.edu
Spatial / Geographical Coverage Location:  Baton Rouge, LA, Pocatello, ID,
Blacksburg VA
Spatial Extent (Lat/Lon):
Temporal Extent (end date): Thu, 06/25/2015
Temporal Extent (start date): Fri, 05/10/2013
Additional Information:
Would you like a DOI for this dataset? Yes
We agree to the NKN Terms of Service: Yes