background:backsub and scimap support in mcmicro

[JiayuanDing100]

Hi,

I am wondering whether background subtraction and scimap are suupported in mcmicro workflow. I saw them both in the output of mcmicro pipeline, but i couldn't see any tutorial about both.

N E X T F L O W ~ version 22.10.6
Launching mcmicro/main.nf [romantic_colden] DSL2 - revision: 4408d4565a
executor > local (11)
[88/4e015d] process > illumination (3) [100%] 7 of 7 ✔
[84/9e72c3] process > registration:ashlar [100%] 1 of 1 ✔
[- ] process > background:backsub -
[- ] process > dearray:coreograph -
[- ] process > dearray:roadie:runTask -
[- ] process > segmentation:roadie:runTask -
[d5/cc7b21] process > segmentation:worker (unmicst-1) [100%] 1 of 1 ✔
[8a/73cd11] process > segmentation:s3seg (1) [100%] 1 of 1 ✔
[3a/270915] process > quantification:mcquant (1) [100%] 1 of 1 ✔
[- ] process > downstream:worker -
[- ] process > viz:autominerva -
Completed at: 06-Apr-2023 03:32:56
Duration : 10m 59s
CPU hours : 0.2
Succeeded : 11

Thanks,

[JiayuanDing100]

Hi,

Great! I saw background subtraction is supported in mcmicro! https://mcmicro.org/parameters/other.html#backsub

However, it is channel based, I am wondering whether it applies for my use case.
Let's say I have 7 cycles, 6 quenches, each one has 4 channels.

For example,
cycle 1: channel 1,2,3,4
quench 1: channel 1,2,3,4,

The rule for background subtraction is like this:
cycle1:channel 1 - quench 1: channel 1
cycle1:channel 2 - quench 1: channel 2

I am wondering whether the. above is supported in mcmicro?

Thanks,

[ArtemSokolov]

Hi @JiayuanDing100,

Yes, any channel can be used as background for any other channel. You can specify these relationships in markers.csv. There is a link to an example in the inputs section of Backsub instructions.

[JiayuanDing100]

Hi @ArtemSokolov ,

Yes, right now I got the idea of how to make backsub work.

I just have a question about the markers.csv, especially the marker_name column:

1. what if i have cycle and quence, each cycle and quench have exactly the same marker proteins. How to differentiate them in marker_name column?

Thanks,

[ArtemSokolov]

Ideally, channel names should be unique. One suggestion is to add a suffix for the quench entries, e.g.,

cycle,marker_name
cycle1,DNA1
cycle1,CD45
cycle1.CD3
cycle1,CD8
quench1,DNA1_q
quench1,CD45_q
quench1,CD3_q
quench1,CD8_q
cycle2,DNA2
...etc.