FISH Data Submission Help

This document describes what we at SGN are prepared to accept for FISH experiment data. Please contact us if you have any questions about these guidelines.

When submitting spreads, the preferred folder layout looks like this:

<chromosome arm>/<BACS on arm>/<spreads for BAC>/<files for spread>

For example, a single spread for a BAC on chromosome 11's long arm would be:

  Tomato_11Q/
  Tomato_11Q/BAC_323E19/
  Tomato_11Q/BAC_323E19/Spread_127/
  Tomato_11Q/BAC_323E19/Spread_127/127_composite.jpg
  Tomato_11Q/BAC_323E19/Spread_127/127_phase.jpg
  Tomato_11Q/BAC_323E19/Spread_127/127_fluorescence.jpg

Other spreads for BAC 323E19 would be beside Spread_127 in the BAC_323E19 folder, other BACs on Tomato's 11Q arm would be beside BAC_323E19 in the Tomato_11Q folder, and BACs on other chromosome arms would be in folders beside Tomato_11Q.

The "Spread" IDs above are arbitrary, but it's best if the submitter uses different IDs for each spread, across their entire FISHing project (we assume that the lab already uses unique identifiers for their experiements, so we can just use those identifiers, for simplicity).

Each spread's folder must have at least the 3 image files. We'll also store any additional per-spread datafiles beside the JPGs in the Spread_NNN folders.

Note If possible, when multiple BACs occur in the same spread, make two distinct folders with a copy of each image in each folder (e.g., have two folders "Spread_15" and "Spread_16", rather than putting the images in a single folder called "Spreads_15_and_16" or something).

For each upload of several spreads, a single spreadsheet with at least the following columns (columns of other data will be ignored):

  BAC ID      Spread   BAC % from kc
  ------      ------   -------------
  323E19       127       86.99
  323E19       128       87.73
  323E19       129       88.27
  323E19       130       90.60
  323E19       131       90.58
  323E19       132       90.01
  323E19       133       86.37
  323E19       134       84.70