deletions | additions       diff --git a/Research_Approach_1_3_4__.md b/Research_Approach_1_3_4__.md  index 20c1c0c..07bb9de 100644  --- a/Research_Approach_1_3_4__.md  +++ b/Research_Approach_1_3_4__.md 
...
####**Aim 1: A global census of microbial eukaryotes found in association with leaves, roots, and sediment of the seagrass Zostera marina.**   _Sampling/Molecular Biology:_ We have, in hand, DNA that has been extracted from 400 samples of _Zostera marina_ leaves, roots, and sediment (ZEN samples). These samples include 3 biological replicates of each sample type. Samples were take from two seagrass beds at each of 22 sites in northern Europe, the Mediterranean, Japan, Australia, Central America, and both coasts of the United States. We have successfully amplified and sequenced the 16S ribosomal RNA genes of the microbial communities in these samples. Analysis of these samples is currently underway, and is showing very exciting results. We have also demonstrated success amplifying, sequencing, and analyzing eukaryotic microbial loci (fungal ITS) from samples prepared in the same way as the ZEN samples. We also have a wealth of metadata associated with each site: 109 measurements of plant traits (including biomass, productivity, and microsatellite loci), water and sediment characteristics, and abundances of invertebrate grazers and predator species. We will use universal microbial eukaryotic primers, targeting the V4 and V9 variable regions of the 18S ribosomal RNA gene. The V4 region will be amplified using the AReuk454FWD1 and TAReukREV3 primers with Illumina adaptors instead of 454 adaptors from \cite{STOECK_2010} and the V9 region will be amplified using the 1391f and EukBr primers according to the earth microbiome project protocols. The V4 and V9 regions are standardly used in such inventories and are known to be informative at the (??? taxonomic) level.   _Bioinformatic Analysis:_ Samples will be multiplexed and the resulting DNA libraries will be sequenced using Illumina MiSeq 250 bp paired end sequencing at the UCD Genome Center. Preprocessing and quality assessment will be performed as described by \cite{Smith_2014}. Sequence analysis will follow a similar worflow to \cite{Korajkic_2015} with sequences analyzed using a modified QIIME, Quantitative Insights Into Microbial Ecology \cite{Caporaso_2010}, workflow using UPARSE \cite{Edgar_2013} to pick open reference operational taxonomic units (OTU’s) at 97% and 99% similarity. Taxonomy will be assigned using the RDP Classifier \cite{Wang_2007} with the SILVA rRNA database \cite{Quast_2012}  and Genbank searches. Phylogenetic trees will be constructed by placing representative sequences for each de novo OTU into the SILVA reference tree with RAxML. The resulting OTU matrix will be rarefied and subsequent data analysis performed in R. Alpha diversity will be calculated using a variety of metrics including the Shannon and Simpson indices to determine intrasample diversity. The beta diversity metric, Bray–Curtis dissimilarity, will be calculated for the OTU matrix to determine intersample diversity. PERMANOVA tests \cite{Anderson_2001}  will be used to determine the significance of sample type (ex. sediment), location and environmental measures (salinity, temperature, etc.) on microbial eukaryotic community composition. To visualize and assess patterns of microeukayotic community structure, a nonmetric multidimensional scaling (NMDS) ordination of the Bray–Curtis dissimilarities will be created. Mantel tests will be performed to determine if correlations exist between the bacterial community composition from the 16S rRNA results and the microeukaryotic community composition we determine here. *Add References - Cassie