Overview

As mentioned in the introduction tutorial as well as the read processing tutorial, read processing can make a huge impact on downstream work. While cutadapt which was introduced in the read processing tutorial is great for quick evaluation or dealing with a single bad sample, it is not as robust as some other trimmers in particular when it comes to removing sequence that you know shouldn't be present but may exist in odd orientations (such as adapter sequences from the library preparation).

Learning objectives:

1. Install trimmomatic

2. Remove adapter sequences from some plasmids and evaluate effect on read quality, or assembly.

Installing trimmomatic

Trimmomatic's home page can be found at this link which includes links to the paper discussing the program, and a user manual. Trimmomatic is far above average for as far as programs go, most will not have a user manual, may not have been updated since originally published, etc. This is one thing that makes it such a good tool. Another sign of its popularity is that while the home page lists information for installing it that is not conda based, searching for the tool on anaconda still reveals this page https://anaconda.org/bioconda/trimmomatic which is a conda installation for the tool. Anytime you have other people in the community working with a tool and sharing resources like this it is a good sign.

As noted in our IVG tutorial, when we dealt with the readseq program, trimmomatic is also a java based program, and like the conda installation of readseq, the conda installation of trimmomatic includes a bash wrapper script around the java invocation. For those interested in how such wrapper scripts are made, last year's tirmmomatic tutorial actually built a trimmomatic wrapper to avoid the java envokation. Using "-version" verify that you have version 0.39 of trimmomatic installed.

Trimming adapter sequences

Example generic command

trimmomatic PE <READ1> <READ2> -baseout Trim_Reads/<basename> ILLUMINACLIP:<FASTAadapterFILE>:4:30:10 MINLEN:30

Breaking down the parts of the above command:

ILLUMINACLIP

<FASTAadapterFILE>

:4:30:10

 MINLEN:30

All of the above has been put together using the trimmomatic manual and experience in our lab given the adapters we use and the library kits we prepare the samples with.

What does this look like in practice you may ask? Read on for some examples of trimming a single sample, or trimming a large number of samples.

Trimming a single sample

Get some data

mkdir -p $SCRATCH/GVA_trimmomatic_1sample/Trim_Reads
cd $SCRATCH/GVA_trimmomatic_1sample
cp $BI/gva_course/plasmid_qc/E1-7* .
cp $HOME/miniconda3/envs/*/share/trimmomatic/adapters/TruSeq3-PE-2.fa .

The ls command should show you 2 gzipped fastq files and a fasta file. You may notice that here that we used a wildcard in the middle of our copy path. This is done so that regardless of what environment you installed trimmomatic into, it should still be able to find the correct adapter file. The downside is that if you have installed it in more than one environment, you may get messages like this:

cp: will not overwrite just-created ‘./TruSeq2-PE.fa’ with ‘/home1/0004/train402/miniconda3/envs/GVA-breseq/share/trimmomatic/adapters/TruSeq2-PE.fa’
cp: will not overwrite just-created ‘./TruSeq2-PE.fa’ with ‘/home1/0004/train402/miniconda3/envs/GVA-multiqc/share/trimmomatic/adapters/TruSeq2-PE.fa’

This is actually not concerning as all 3 of these files are from the same source, are identical not just in name, but also in content, and we don't care which of the files we use.

Trim the fastq files

The following command can be run on the head node. Like with FastQC if we are dealing with less than say 1-2Million reads, it is reasonable to run the command on the head node unless we have 100s of samples in which case submitting to the queue will be faster as the files can be trimmed all at once rather than 1 at a time. Use what you have learned in the class to determine if you think this command should be run on the head node. (this was covered in more detail in the first part of the evaluating and processing read quality tutorial.)

zgrep -c "^+$" *.fastq.gz

trimmomatic PE E1-7_S187_L001_R1_001.fastq.gz E1-7_S187_L001_R2_001.fastq.gz -baseout Trim_Reads/E1-7.fastq.gz ILLUMINACLIP:TruSeq3-PE-2.fa:4:30:10 MINLEN:30

Evaluating the output

The last 2 lines of text you get should read:

Input Read Pairs: 6891 Both Surviving: 2391 (34.70%) Forward Only Surviving: 1844 (26.76%) Reverse Only Surviving: 2644 (38.37%) Dropped: 12 (0.17%)
TrimmomaticPE: Completed successfully

2nd to last line tells us:

1. we had 6891 total reads
2. 34.7% of reads both R1 and R2 were long enough to be kept after trimming
3. 26.76% of reads and 38.37% of reads only 1 of the reads were long enough and/or not a complete duplicate of the other read
4. only 0.17% of reads were discarded  for both R1 and R2. This is a rough estimate of adapter dimer being present in the sample.

From the last line alone we would believe things worked correctly. Unfortunately if we run the command with an adapter file that we don't have access to (say you have a typo in the name of the adapter file) our full output would look like this:

TrimmomaticPE: Started with arguments:
 E1-7_S187_L001_R1_001.fastq.gz E1-7_S187_L001_R2_001.fastq.gz -baseout Trim_Reads/E1-7.fastq.gz ILLUMINACLIP:TruSeq3-PE-2.fa:4:30:10 MINLEN:30
Using templated Output files: Trim_Reads/E1-7_1P.fastq.gz Trim_Reads/E1-7_1U.fastq.gz Trim_Reads/E1-7_2P.fastq.gz Trim_Reads/E1-7_2U.fastq.gz
java.io.FileNotFoundException: /scratch/0004/train402/GVA_trimmomatic_1sample/TruSeq3-PE-2.fa (No such file or directory)
	at java.io.FileInputStream.open0(Native Method)
	at java.io.FileInputStream.open(FileInputStream.java:195)
	at java.io.FileInputStream.<init>(FileInputStream.java:138)
	at org.usadellab.trimmomatic.fasta.FastaParser.parse(FastaParser.java:54)
	at org.usadellab.trimmomatic.trim.IlluminaClippingTrimmer.loadSequences(IlluminaClippingTrimmer.java:110)
	at org.usadellab.trimmomatic.trim.IlluminaClippingTrimmer.makeIlluminaClippingTrimmer(IlluminaClippingTrimmer.java:71)
	at org.usadellab.trimmomatic.trim.TrimmerFactory.makeTrimmer(TrimmerFactory.java:32)
	at org.usadellab.trimmomatic.Trimmomatic.createTrimmers(Trimmomatic.java:59)
	at org.usadellab.trimmomatic.TrimmomaticPE.run(TrimmomaticPE.java:552)
	at org.usadellab.trimmomatic.Trimmomatic.main(Trimmomatic.java:80)
Quality encoding detected as phred33
Input Read Pairs: 6891 Both Surviving: 6891 (100.00%) Forward Only Surviving: 0 (0.00%) Reverse Only Surviving: 0 (0.00%) Dropped: 0 (0.00%)
TrimmomaticPE: Completed successfully

Considering this output, we also want to be very suspicious of a trimming that results in 100% of the reads doing anything.

When we interrogate the Trim_Reads folder with ls we see 4 files:

1. _1P
2. _2P
3. _1U
4. _2U

1/2 represent read 1 and read2 ... P/U represent paired and unpaired reads. They are kept separate as many programs require R1 and R2 files to be the same length when being used as input.

Trim all the samples from the multiqc tutorial

As mentioned above, if you have already done the multiqc tutorial, you can use your new trimmomatic command to remove the adapter sequences from all 544 samples.

Get some data

mkdir -p $SCRATCH/GVA_trimmomatic_multiqcSamples/Trim_Reads
cd $SCRATCH/GVA_trimmomatic_multiqcSamples
cp -r $BI/gva_course/plasmid_qc/ Raw_Reads/
cp $HOME/miniconda3/envs/*/share/trimmomatic/adapters/TruSeq3-PE-2.fa .

 Again we may see additional output about not overwriting files if we have trimmomatic installed on multiple environments. The ls command will now show 2 directories, and a fasta file.

Trim the fastq files

Just as we used a for loop to set up a set of FastQC commands in the multiqc tutorial, we can use a similar for loop to generate a single file with 272 trim commands for the 544 total files.

The following command will pair all R1 reads in the Raw_Reads folder with its R2 pair, determine the base name, and generate a command to trim the file 

for r1 in Raw_Reads/*_R1_*.fastq.gz; do r2=$(echo $r1|sed 's/_R1_/_R2_/'); name=$(echo $r1|sed 's/_R1_001.fastq.gz//'|sed 's/Raw_Reads\///'); echo "trimmomatic PE $r1 $r2 -baseout Trim_Reads/$name.fastq.gz ILLUMINACLIP:TruSeq3-PE-2.fa:4:30:10 MINLEN:30";done > trim_commands

Use the head and wc -l to see what the output is and how much there is of it respectively.

Again as we discussed in the multiqc tutorial, running this number of commands is kind of a boarder line case, there are not a lot of total reads, but there are a large number of samples so we are likely to benefit from not being run on an idev node. 

cp /corral-repl/utexas/BioITeam/gva_course/GVA2021.launcher.slurm trim.slurm
nano trim.slurm

Again while in nano you will edit most of the same lines you edited in the in the breseq tutorial. Note that most of these lines have additional text to the right of the line. This commented text is present to help remind you what goes on each line, leaving it alone will not hurt anything, removing it may make it more difficult for you to remember what the purpose of the line is

The changes to lines 22 and 23 are optional but will give you an idea of what types of email you could expect from TACC if you choose to use these options. Just be sure to pay attention to these 2 lines starting with a single # symbol after editing them.

Line 27 assumes you named your breseq environment GVA-breseq in the earlier tutorial. 

Again use ctl-o and ctl-x to save the file and exit.

sbatch trim.slurm

The job should take less than 10 minutes once it starts if everything is working correctly, the showq -u command can be used to check for the job finishing.

Evaluating the output

ls Trim_Reads | wc -l
grep -c "TrimmomaticPE: Completed successfully" Queue_job.o*
grep -c "100.00%" Queue_job.o*

The above 3 commands are expected to show 1088 and 272 and 0 respectively, if you see other answers it suggests that something went wrong with the trimming command itself. If so remember I'm on zoom if you need help looking at whats going on. The most common error that I expect will be that you ran out of ram by trying to process too many samples at once (such as if you used a -n 68 option in your .slurm file). 

Beyond the job finishing successfully, the best way to evaluate this data would actually be to move back to the multiqc tutorial and repeat the analysis there that was done on the raw files for the trimmed files here.

Personal experience

The for loop above focuses just on generating the trim commands. In my experience that is only half of the job, the other half is capturing the individual outputs so you can evaluate how many reads were trimmed in what way for each sample. Perhaps you will find this command helpful in your own work: 

mkdir trimLogs; mkdir Trim_Reads; for r1 in Raw_Reads/*_R1_*.fastq.gz; do r2=$(echo $r1|sed 's/_R1_/_R2_/'); name=$(echo $r1|sed 's/_R1_001.fastq.gz//'|sed 's/Raw_Reads\///'); echo "trimmomatic PE $r1 $r2 -baseout Trim_Reads/$name.fastq.gz ILLUMINACLIP:TruSeq3-PE-2.fa:4:30:10 MINLEN:30 >& trimLogs/$name.log"; done > trim_commands

After the job completes, the following command is useful for evaluating its success:

echo -e "\nTotal read pairs:";wc -l trim_commands;echo "Successful trimmings:"; tail -n 1 trimLogs/*|grep -c "TrimmomaticPE: Completed successfully";echo "Potential trimming errors:"; cat trimLogs/*|grep -c "100.00%"

This gives me a quick readout of if all of the individual commands finished correctly.

Further, if you were to use this for loop rather than the one listed in the tutorial above, you would see each sample generates its own log file in the the trimLogs directory that when investigated with cat/more/less/tail/nano is identical to the output you saw when you trimmed a single sentence allowing you to figure out how different samples are being processed.

Optional next steps:

1. Consider moving back over to the multiqc tutorial use multiqc to determine well the trimming worked. 
2. The reads could then further be assembled in the Spades tutorial as they are all plasmid samples.

Return to GVA2021