4 Sequencing Data Best Practices

Where to put data, how does it link to metadata? How to manage old data?

SampleDB and its Data Manager Utility is currently still under construction 🚧. For now, please store raw sequencing data in the Mines following the following guidelines:

4.1 SeekDeep filename guidelines

experimentIdentifier-micronics-plate-well-rep-commentInfo

  1. experimentID (of some kind)
  2. micronics barcode (use NC# for negative controls)
  3. plate name
  4. well position on plate (use A1 not A01)
  5. replicate identifier (use rep1, rep2, etc.)
  6. (optional) comments, e.g. experimental conditions or other info

Rules:

  • Filename must consist of 5 “parts” separated by dashes “-”
  • An optional 6th field is reserved for comments and other information
  • Do not introduce new dashes “-”

From micronics, we can recover date, study_code, identifier, etc. from SampleDB

Example sample filenames:

PBC003-8041439280-Plate2-B7-rep1
S3A003-8041439281-SMT4-A5-rep2
JDH002-8041439282-plate6-C12-rep1

Example sample filenames with optional comment field:

PBC003-8041439280-Plate2-B7-rep1-commentInfo
S3A003-8041439281-SMT4-A5-rep2-commentInfo
JDH002-8041439282-plate6-C12-rep1-commentInfo

Example control filenames:

PBC003-NC1-Plate7-E9-rep1
S3A003-NC3-plate9-F11-rep2
JDH002-NC10-PBCSMT4-G9-rep2

4.2 Raw sequencing data 🚧

Please store raw sequencing data in the mines in the following format: sequencing_data/YYYYMMDD/some_filename.gz

4.2.1 Folder names 🚧

Date: The YYMMDD date should be the sequencing run start date

Biohub
YYMMDD_Sequencer_RunNumber_FlowCellSerialNumber
Example folder name:
221006_VH00444_149_AAATC37HV

In-house: Make the folder name manually and be diligent about removing or renaming erroneously demuxed data. Here we emulate the folder naming structure used by Biohub.
YYMMDD_Sequencer_RunNumber_FlowCellSerialNumber
Example folder name:
221006_MiniSeqRR_1