Skip to content

Tutorial 2: Inspect raw reads on a phone

This tutorial takes one sequencing run from a public archive to a full quality report, using only a phone. You will:

  1. Find the run in the Sequence Read Archive.
  2. Download its FASTQ files.
  3. Summarize the reads in Complete mode.
  4. Read the report: quality, composition, duplication, and adapters.
  • A phone or tablet with Android 8.0 or newer, or iOS or iPadOS 15.0 or newer.
  • SEGUI v1.2.0 or newer. See GUI installation.
  • About 500 MB of free storage: 238 MB for the download and the same again for SEGUI's working copy.
  • Wi-Fi. The two files add up to 238 MB.

The data are the raw reads of one sample from Esselstyn et al. 2017, a study of the placental mammal tree based on ultraconserved elements (UCEs). The sample is a short-tailed gymnure, Hylomys suillus, from the Field Museum (FMNH 168652).

RunSRR6053059
BioProjectPRJNA390442
LibraryUCE target enrichment, paired-end
InstrumentIllumina HiSeq 3000
Reads1,761,846 pairs, 125 bp each
FilesSRR6053059_1.fastq.gz, 119 MB, and SRR6053059_2.fastq.gz, 119 MB

Reads submitted to GenBank's Sequence Read Archive (SRA) get a run accession that starts with SRR. A paper lists either the runs or a BioProject that groups them.

  1. Open the SRA record of SRR6053059 in your browser.
  2. Check the record: the organism is Hylomys suillus, the layout is PAIRED, and the run has 1.8M spots and 440.5M bases. A spot is one read pair.
  3. Note the run accession, SRR6053059. The same accession finds the run in every archive that mirrors the SRA.
  1. In Chrome, tap SRR6053059_1.fastq.gz. The download starts.
  2. Tap SRR6053059_2.fastq.gz.
  3. To follow the progress, open the Chrome menu and select Downloads.
  4. Chrome saves both files to the Downloads folder. Wait for both to finish. You do not need to open or extract them.
SRA record of SRR6053059 in Chrome on Android with the Hylomys suillus library and 1,761,846 spots
The SRA record of the run. Zoom in to read it.
Chrome Downloads page on Android with SRR6053059_1.fastq.gz and SRR6053059_2.fastq.gz in progress
Both files downloading in Chrome.
  1. Open SEGUI. On the home screen, select Summarize genomic reads.

  2. In the Input card, select Browse, then Browse files.

  3. Pick both FASTQ files:

    In the system file picker, open the menu and select Downloads. Touch and hold SRR6053059_1.fastq.gz, tap SRR6053059_2.fastq.gz, then tap Select.

  4. Wait while SEGUI copies the files. The input card then shows 2 files selected.

  5. Leave Format on Auto.

SEGUI home screen on Android with Summarize genomic reads at the top of the Genomics tools
The tool on the home screen.
SEGUI Select files sheet with Browse files and Browse folder
Choose Browse files.
Android file picker with SRR6053059_1.fastq.gz and SRR6053059_2.fastq.gz selected in the Download folder
Select both FASTQ files.
  1. In the Output card, enter a Directory name, such as read-summary.
  2. Enter a Prefix, such as hylomys. It starts the name of each output file.
  3. Leave Summary Mode on Complete. Only this mode writes the read report.
  4. Leave Duplication sketch size on 100k and Quality score offset on Auto.
  5. Select Summarize.

The run takes a few seconds. SEGUI then opens the Workbench tab, with 2 input files and 4 output files.

SEGUI Summarize genomic reads output card with directory name read-summary, prefix hylomys, and Complete summary mode
The output settings before the run.
SEGUI running the read summary with a progress bar at 0 of 2 files and a Cancel button
Progress while SEGUI reads the files.
SEGUI workbench after the read summary with 2 input files, 4 output files, and the Report card
The workbench after the run, with the Report card.

The output files are:

  • hylomys_default-read-summary.csv: one row of statistics per file.
  • hylomys_read-pos-summary.zip and hylomys_read-dist-summary.zip: tables per read position and distributions for each file.
  • hylomys_read-report.json: the read report.

Select View report on the Report card. The report shows one file at a time. The two files are the two directions of the same read pairs, so inspect both.

The Summary tab lists both files with their read counts. Scroll down to the checks of the open file. Each one is marked pass, warn, or fail.

CheckSRR6053059_1SRR6053059_2
Q30 bases85.7%83.2%
Mean quality33.332.6
N content0.0%0.1%
Duplication76.0%73.7%
Adapters0.4%0.4%
Poly-G0.1%0.1%
Overrepresented0.2%0.2%

Quality, N content, adapters, and poly-G pass. Duplication fails and overrepresented sequences get a warning. Tap a check to see what it measures and its thresholds.

To open the other file, use the arrows next to the file name, or the list button to see both files with their reads and Q30 bases. For every statistic side by side, tap the settings button above the bars and select Table.

SEGUI read report Summary tab with both files at 1.8M reads and the checks of SRR6053059_1
The Summary tab with the checks.
SEGUI Duplication check sheet with 76.0% marked fail against the pass, warn, and fail ranges
A check with its thresholds.
SEGUI sample list sheet with SRR6053059_1 at Q30 85.7% and SRR6053059_2 at Q30 83.2%
The sample list switches files.

Open the Quality tab. The per-base quality chart plots the quality score at each position of the read, over green, yellow, and red zones for good, reasonable, and poor quality.

  • The median stays at 37 from position 6 to 124, well inside the green zone.
  • The mean falls slowly, from about 35 near the start to 30 at position 120, then drops to 23 at the last base.
  • The shaded band shows the 10th percentile. It reaches a score of 2 at position 115 in the first file and at position 107 in the second.

So most bases are good, but a share of reads have poor ends, and the second file is slightly worse than the first. Both are typical of Illumina paired-end runs. The bands move in steps because this instrument reports only a few quality values.

SEGUI per-base quality chart of SRR6053059_1 with a median of 37 and a mean falling toward the read end
Per-base quality of the first file.
SEGUI per-base quality chart of SRR6053059_2 with the low quality band starting earlier in the read
The second file loses quality earlier.

Scroll down to Per-read quality, which compares both files. Most reads have a mean quality above 30. In total, 82,824 reads of the first file and 118,992 of the second have a mean quality below 20, or 4.7% and 6.8%.

Open the Composition tab.

  • GC content: the reads average 46.3% GC, and their distribution follows the theoretical curve closely. A second peak would point to contamination.
  • Base composition: the four lines are flat after the first few bases, as expected of a random library.
  • N content: the sequencer could not call 0.01% of the bases in the first file and 0.13% in the second.

Touch a chart to read its values, and pinch to zoom.

SEGUI GC content chart of SRR6053059_1 with the observed distribution close to the theoretical curve
GC content against the theoretical curve.

Duplication, adapters, and overrepresented sequences

Section titled “Duplication, adapters, and overrepresented sequences”
  • Duplication: 76.0% of the reads in the first file repeat another read. This fails the check, but it is common in target enrichment libraries, which sequence a small part of the genome many times after PCR. Remove duplicates before calling variants, or let the assembler handle them.
  • Adapters: the Illumina Universal adapter is found in 0.4% of the reads, mostly from position 100 onward, where short fragments run out. The line is low, but it rises, so trim the reads before assembly.
  • Overrepresented: one sequence makes up 0.2% of each file. Both start with GATCGGAAGAGC, which is Illumina adapter sequence. These are reads of adapters with almost no insert, and trimming removes them. Use the copy button to search for a sequence elsewhere.
SEGUI duplication chart of SRR6053059_1 showing 76.0% of 1,761,846 reads repeat another read
Duplication of the first file.
SEGUI adapter content chart of SRR6053059_1 with the Illumina Universal line rising to 0.4% near the read end
Adapter content along the read.
SEGUI overrepresented sequences table of SRR6053059_1 with one sequence starting with GATCGGAAGAGC
The overrepresented sequence.

The Length tab has little to show here: every read is 125 bp, because the reads are untrimmed.

StatisticSRR6053059_1SRR6053059_2
Reads1,761,8461,761,846
Bases220,230,750220,230,750
Read length125 bp125 bp
GC content46.3%46.4%
Mean quality33.332.6
Q20 bases90.7%88.5%
Q30 bases85.7%83.2%
Reads with mean quality below 2082,824118,992
Duplication76.0%73.7%
Reads with adapters0.4%0.4%

The run is good enough to use: both files have the same number of reads, more than 83% of the bases reach Q30, and there is no sign of contamination. Before assembly, trim the adapters and the low quality read ends.

To keep the results, select the back arrow, then Quick Share on the Workbench tab. See exporting output. For every chart and control of the report, see the read report guide.

The same summary runs from the command line with segul read summary and --mode complete, and gives the same numbers.

This material is based upon work supported by the National Science Foundation under grant numbers: DEB-1754393 and DEB-2244754. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.