NGS LIMS features

A generic LIMS models one sample producing one result. A sequencing sample produces a library, the library joins a pool, the pool occupies part of a flow cell, the run produces reads that are demultiplexed back to the library, and a pipeline turns those reads into alignments and calls. Any of those steps can be the reason a result looks wrong.

When the intermediate steps live in a shared spreadsheet, troubleshooting means asking who remembers. When they live in structured records, the failing step is usually obvious within a few minutes: a low-input library, a duplicated index, an underloaded flow cell, or a pipeline run against the wrong reference build.

Each library should carry, at minimum:

• Parent sample and extraction batch so DNA/RNA quality can be traced back.
• Kit, lot, and protocol version — the fastest way to spot a bad kit lot across runs.
• Input mass and fragmentation method, with the measured concentration used to compute it.
• PCR cycle count, the usual explanation for duplicate-heavy libraries.
• Final concentration and fragment size distribution from a fluorometer and a bioanalyzer trace.
• Operator and date, so protocol drift between people is visible.

Indexing is where an NGS LIMS earns its keep. The system should hold the index kit, the i7 and i5 sequences assigned to each library, and reject a pool containing colliding or insufficiently distant index pairs before the flow cell is loaded rather than after.

Pooling records should store the molarity of each library, the volume contributed, the expected read share, and the resulting pool concentration. After demultiplexing, comparing expected read share against observed read share turns a vague "the run was uneven" into a per-library number you can act on.

The run record ties everything together and should include:

• Flow cell ID, instrument, and chemistry or kit version.
• Run type and read configuration (single or paired end, read and index lengths).
• Loading concentration and any spike-in control, with its expected percentage.
• The pool loaded on each lane and the libraries inside it.
• Run start and completion times, plus the operator.
• The demultiplexing sample sheet actually used, kept with the run rather than regenerated later.

Storing these as fields rather than attached PDFs is what allows trending: a slow decline in %Q30 across ten runs is a maintenance signal, invisible if every metric lives in a separate report file.

The analysis half of an NGS workflow needs the same discipline as the bench half. Record the pipeline name and version, the container or environment used, the reference genome build and annotation release, the parameters, and the location of inputs and outputs.

With those five fields, a result is reproducible; without them it is an opinion. They also make it possible to answer the question reviewers actually ask — "was this cohort processed with the same pipeline?" — without re-reading anyone's notes.

Sanger and targeted PCR work belongs in the same record set. Trace files, primer pairs, annealing temperatures, expected versus observed band sizes, and BLAST identifications sit naturally next to NGS runs when both describe the same samples.

Sequencing data is often human-derived, shared across institutions, and retained for years. An NGS LIMS should provide per-project access control with distinct roles for students, supervisors, and lab managers, an audit trail of who changed which record, and the ability to share a single project with an external collaborator without exposing the rest of the database.

1. Are libraries, pools, and runs separate record types, or squeezed into one "sample" table?
2. Can the system detect index collisions before a flow cell is loaded?
3. Does it store per-stage QC metrics as queryable fields rather than attachments?
4. Can pipeline version, reference build, and parameters be recorded against a run?
5. Does it link raw files to the library and run that produced them?
6. Can a project be shared with a collaborator without giving away everything else?
7. Can you export every table in an open format the day you decide to leave?
8. Can a new student learn it in an afternoon? Unused systems track nothing.

ResearchHub Intelligence is a research workspace for academic and molecular biology labs rather than an industrial sequencing facility. It tracks projects and samples with storage and location, keeps sequencing runs and PCR/gel records against those samples, parses .ab1 trace files for read length, mean quality, Q20, and GC content, screens sequences for antimicrobial resistance genes, and runs BLAST-based species identification.

Around that sit the parts most sequencing labs otherwise export for: flexible worksheets with Excel and CSV import, descriptive statistics, t-tests, ANOVA and regression, qPCR quantification with standard curves, charts, literature and citation management, and report, poster, and manuscript drafting.

What is an NGS LIMS?

An NGS LIMS is a laboratory information management system built for next-generation sequencing workflows. On top of ordinary sample tracking it records library preparation, indexes and barcodes, pooling, flow cell and run metadata, and links every downstream FASTQ, BAM, or variant file back to the sample it came from.

How is it different from a general-purpose LIMS?

A general LIMS tracks a sample from collection to result. NGS adds several layers between those two points — extraction, library prep, indexing, pooling, sequencing run, demultiplexing, and pipeline analysis — and each layer has its own QC metrics. An NGS LIMS models those steps as first-class records instead of free-text notes.

What library preparation data should be tracked?

Kit and lot number, fragmentation method, adapter and index pair, PCR cycles, input mass, final library concentration and fragment size, and the operator and date. Index pairs matter most: recording them prevents index collisions when libraries are pooled onto one flow cell.

What flow cell information matters?

Flow cell ID, instrument and chemistry, run type and read length, loading concentration, cluster density or occupancy, %Q30, total reads passing filter, and the list of pooled libraries with expected read share. These fields explain nearly every underloaded or low-quality run after the fact.

How does a LIMS integrate with bioinformatics pipelines?

By storing the pipeline name and version, the reference genome build, the parameters used, and the location of inputs and outputs against the run record. That is what makes an analysis re-runnable a year later and what reviewers ask for when a variant call is questioned.

Do small academic labs need an NGS LIMS?

A lab running a handful of runs per year can survive on spreadsheets, but index assignments, pooling ratios, and pipeline versions are exactly the details that get lost between students. A lightweight system that records them is usually enough — a full industrial LIMS rarely is.

Can ResearchHub Intelligence handle sequencing data?

Yes. It tracks samples, sequencing runs, and PCR/gel records, parses .ab1 trace files for quality metrics, screens sequences for resistance genes, and runs BLAST-based species identification — with projects, worksheets, statistics, and reports around them.