LRII Bioinformatics

Assign reads and otus to samples:

vsearch -sintax OTUFILE -db REFERENCEDATABASE -tabbedout OUTPUT -sintax_cutoff 0.8 -strand both -threads 32

Assign taxonomy

Output:

Reading file /project/microbiome/users/grandol1/ref_db/gg_16s_13.5.fa 100%  

1769520677 nt in 1262986 seqs, min 1111, max 2368, avg 1401

Counting k-mers 100% 

Creating k-mer index 100% 

Classifying sequences 100%   

Classified 4038 of 4042 sequences (99.90%)

Convert into useful form:

LowReadII Analysis

This experiment intended to answer the questions below.

Questions:

1. How can we best accommodate low concentration input DNAs?

2. Concentrate input DNAs?

3. Concentrate/normalize low yield products based off qPCR?

4. Both?

5. Just put more volume of original DNA in?

The questions actually probed are 1, 5, and 3. Though, one could argue that 2 and 5 are similar enough, without doing a cleanup of the DNA as well.

Based off 2-step prep sample sequence outputs, we selected 4 average read samples and 8 low read samples. We used our 1-step prep for this experiment though. All samples were normalized to 10 ng/ul or less. 2 ul of each average read sample was used for the 1-step PCR. The low read samples were repeated 3 times at different volumes (2ul, 4ul, and 8ul). 2 ul of 10ng/ul Zymo Mock Community (consisting of Listeria monocytogenes - 12%, Pseudomonas aeruginosa - 12%, Bacillus subtilis - 12%, Escherichia coli - 12%, Salmonella enterica - 12%, Lactobacillus fermentum - 12%, Enterococcus faecalis - 12%, Staphylococcus aureus - 12%, Saccharomyces cerevisiae - 2%, and Cryptococcus neoformans - 2%) was used as a positive control in quadruplicate, and 2 ul of ultra pure water was used as a negative control. After pcr and cleanup, 2 ul from each of these was pooled. All samples have "_2", "_4", or "_8" appended to the sample name to indicate DNA volume used in PCR.

The whole above setup was repeated with the same samples with the intention of addressing question 3. After pcr and cleanup, these preps were all quantified via qPCR and various volumes were combined to have each prep equimolar in the pool. These sample names have "_N" appended to them for normalized.

Both pools were then quantified via qPCR, adjusted to 1 nMolar and combined in equal amounts to maintain 1 nMolar concentration. After dilution to 50pM, this was the pool used for sequencing on the iSeq

Read in otu table. Transpose it. Remove added values to front of sample names (mid assignments). Sort by samplename and remove extraneous text, "centroid="

Open

Looking at total reads, qPCR normalization and DNA volume used does not appear to have a consistent impact. Why does there appear to be a disconnect between qPCR results and Reads? We separated out each sample and compared it to itself under the different conditions.

Open

There still appears to be a disconnect, though from 2 to 4 ul there is a more consistent increase for the non-normalized. Reads do not consistently go up with increased DNA starting Volume. So, is there a problem with our sample setup? Is our demux key incorrect? Are blanks largely ISD?

[1] "Blank1_2" "Blank1_2" "Blank2_2" "Blank2_2" "Blank3_2" "Blank3_2" "Blank4_2" "Blank4_2"
[9] "SAG_S30P2R_8" "SAG_S30P2R_8" "SAG_S38P1R_8" "SAG_S38P1R_8" "Blank1_2_N" "Blank1_2_N" "Blank2_2_N" "Blank2_2_N"
[17] "Blank3_2_N" "Blank3_2_N" "Blank4_2_N" "Blank4_2_N" "SAG_S15P4R_8_N" "SAG_S38P1R_8_N" "SAG_S38P1R_8_N"

All of the 8 ul iterations of SAG_S38P1R appear to be largely ISD. All of the samples appearing in the Blank range are 8 ul samples. Does too much DNA or too much inhibitor(s) in DNA favor the ISD?

Blanks look good excluding one suspect at only 67% ISD. what about the Mock Community? Is Mock Community largely the 8 expected taxa? (Listeria monocytogenes, Pseudomonas aeruginosa, Bacillus subtilis, Escherichia coli, Salmonella enterica, Lactobacillus fermentum, Enterococcus faecalis, Staphylococcus aureus)

[1] NA "f:mitochondria" "f:Lactobacillaceae" "f:Pseudomonadaceae" "f:Enterobacteriaceae"
[6] "f:Bacillaceae" "f:Staphylococcaceae" "f:Listeriaceae" "f:Enterococcaceae" "f:[Chthoniobacteraceae]"

So, there are some contaminants that should not be in the Mock Community.

[1] "otu31" "otu18" "otu51" "otu1443" "otu1766" "otu2540" "otu3273" "otu3278"

Are these contaminants present in the blanks? 4 of them are

The Mock Community appears to be as expected with some contamination. How do the samples look? Could any of them be MC or blanks remembering that some 8 ul DNA samples are showing high ISD.

Less than 5% of the reads for the non control samples are from the contaminants present in the blanks. How does this carry over per sample?

*PercDiff is the percent of reads coming from the contaminants in the blanks.

Graph is all samples having more than 10% of the reads from blank contaminants.

We should repeat the normalization after qPCR part of this experiment because our qPCR machine needed recalibration and the redone results are much different for the "normalized" samples. Increasing DNA into the PCR rxn does not seem to have the desired impact though.