Help

Try an example

Load a complete small library into the submission form. Its reference genome and assay are selected for you; library structure stays in Auto mode. Review the settings, optionally add an email for notification, confirm the data policy, and select Start analysis.

These research examples contain no patient samples. The downloads retain the reads used for the saved reports. Updated software, reference resources or settings can produce different values when you rerun them.

Input files

Upload basecalled ONT reads as .fastq, .fq, .fastq.gz or .fq.gz. POD5 and FAST5 are not accepted; convert them to FASTQ with your basecalling workflow first. Renaming an extension does not convert a file.

Use ordinary four-line FASTQ: a unique read name beginning with @, a DNA sequence, a + line, and one quality character per base. Read names may contain letters, numbers, dots, underscores, colons, equals signs, plus signs and hyphens. Space- or tab-separated ONT header metadata is accepted.

@example_read_1
ACGTACGT
+
IIIIIIII

This short record illustrates the file format; use the downloadable libraries above for an analysis.

Each FASTQ is analyzed independently. A file may contain multiple index-defined sublibraries. For multiple files, choose settings for each or use Apply to all to copy the reference and assay. Library-structure settings have their own apply control.

Use filenames of at most 180 characters, starting with a letter or number; English letters, numbers, spaces, dots, underscores and hyphens are accepted. Sample names must start with a letter and use only letters, numbers, underscores and hyphens, up to 48 characters.

Choose settings

Reference genome

Select the assembly that matches the library. An incorrect assembly can reduce mapping and distort coverage or contact metrics. Availability is shown in the submission form.

Library assay

Chromatin covers libraries such as CUT&RUN, CUT&Tag, ChIP and ATAC, with insert-size and coverage profiles. Micro-C reports contact pairs, ligation structure and digestion evidence.

Library structure

Auto detects supported P5/P7, index and adapter patterns during analysis. Choose Customize for a known layout: set an adapter preset (TruSeq or Nextera), override adapter sequences, specify index lengths, or add fixed-length UMIs at their physical positions. Each UMI can be 1–64 bp; its position and length are independent. Drag a UMI or use its position menu to move it. The Read 1/Read 2 arrows indicate sequencing direction, not physical sequence blocks.

UMI extraction

UMIs are extracted at configured offsets on the P5-to-P7 strand. Insert sequences used for mapping exclude configured UMIs; full-library lengths include them. ONT insertions and deletions can affect extraction. UMI configuration does not turn on UMI-based deduplication.

Alignment method

Leave the method at Default unless you need an available alternative. Chromatin uses Bowtie2 by default. Micro-C normally uses BWA-MEM2; some assemblies use an explicitly configured BWA-MEM alternative on memory-limited servers. The form lists the methods currently available.

Screening

Mycoplasma and spike-in/foreign-sequence screens depend on the installed reference panels. They are supporting evidence, not definitive contamination tests. A negative screen does not exclude contamination; indicate whether a spike-in was intentionally added.

Custom BED regions

For Chromatin, optionally upload a BED file of up to 4 MiB, using the selected assembly and BED's zero-based, half-open coordinates. Confirm the assembly. The report also allows browser-based profiles from a BED file, up to 100,000 regions per profile. Select the file inside a folder, not a folder whose name ends in .bed.

Advanced settings

Normally keep the defaults. Expected physical index pairs use one name i5 i7 triple per line. Leave this blank for automatic discovery. Thresholds are local screening guides, not universal biological pass/fail criteria.

Read the report

Start with the total-library summary, then compare index-defined sublibraries where present. Expand the interpretation and methods sections beside the plots for the definitions and denominators used in that report.

Library structure and indices

Check adapter completeness and the physical i5/i7 pairs. Unexpected pairs or many incomplete structures can indicate mixed libraries, incomplete molecules or a layout that needs custom settings. Inspect the evidence before assigning a biological cause.

Full-library profile vs. raw ONT molecule size

The count-based full-library profile counts reconstructed library units. Raw ONT size measures each sequenced molecule. One ONT molecule can contain multiple library units, so these distributions need not agree. Length-weighted profiles emphasize longer molecules and answer a different question from counts.

Mapping and Chromatin profiles

Read mapping rates together with reference choice, reconstructed insert length and usable-read counts. Gene and custom-region metaplots describe the distribution of retained signal; normalization and low depth affect their interpretation.

Micro-C contacts

Use the contact funnel and pairtools plots to see how many candidates remain after mapping and filtering. Cis/trans proportions, contact-distance profiles, ligation evidence and MNase-related sizes describe different stages of the library. A percentage based on few retained contacts can be unstable.

Sublibraries and QC flags

Compare each sublibrary's usable-read count as well as its rates. Colored flags use configured reference cutoffs; they do not certify experimental success or failure. Very shallow samples may support only descriptive results.

Open the Chromatin report or Micro-C report to explore the same plots and controls available in a new result. Download the report bundle to keep a local copy of its pages, figures and tables.

Limits, delivery and privacy

The public service currently accepts up to 16 files per submission, 50 MB per file, 250 MB after gzip expansion, and 250,000 reads per file. The upload form displays the current server limits. Larger inputs can be analyzed with the local pipeline.

Cookies are not required to submit files, view reports or download results. Save the private task link as soon as it appears. It shows queued, running, finished or failed status and allows you to return later. Email is optional. If you provide an address, one summary email is sent after all files in a submission have finished processing. Without email, save your private task links to collect your results; lost links cannot be recovered here. Anyone with a private link can access that task; share it only with people you trust.

Uploaded reads and calculation files are removed after processing. Reports are retained for 7 days, unless you delete them earlier. Download what you need before expiry. Deletion is permanent.

Submit only sequencing files you are permitted to process. Do not upload human patient data, including de-identified patient data, or other sensitive personal data.

Common problems

Invalid FASTQ or duplicate read identifier

Check that the file is complete, has four lines per record, has matching sequence/quality lengths, and contains unique read names. A valid gzip download should decompress without an error.

Low mapping or few reconstructed units

Check the assembly and assay first, then adapter/index settings, UMI positions and lengths, and the length/quality distribution. Compare sublibraries separately. Low mapping alone cannot identify its cause.

The job is queued or the service is busy

Keep your task link and return later. Queued work starts when capacity is available. If an upload was rejected, wait briefly before retrying; avoid repeatedly resubmitting the same file.

A report link has expired or is incomplete

Reopen the full private task link from your email or saved links. It provides fresh access to the report and downloads while the report is retained. A temporary figure or report URL is not a replacement for that private task link.

A figure did not load

Reload the report after a short wait. You can also open the figure directly or use the downloaded report bundle. If the problem persists, contact us with the sample name and the failed figure, without posting private links publicly.

Expired task or missing access key

An expired task may have reached the retention limit or been manually deleted. A report URL alone does not always include its access key; use the complete private task link from your email or saved links.

No email received

If you left the optional email field blank, no email will be sent. If you provided an address, check spam and wait for all files in the submission to finish. The saved private task link remains the direct way to check status and retrieve available reports.

For help, email [email protected] or use GitHub Issues. Do not post private task links or sequencing data in a public issue.

Liu, H., & Kim, J. J. (2026). ONT-LibQC: rapid quality control of chromatin assay libraries using shallow nanopore sequencing. bioRxiv [Preprint]. https://doi.org/10.64898/2026.10.19.760194