TRaceR

Pipeline

This is a Kerblam! data analysis project. Provided that you have the right input data structure, just run the tracer :: Multiparametric Calcium Trace Analyzer workflow through

kerblam run tracer

to get the final output.

Run

kerblam run tracer 2>&1 | tee log.txt

to also locally save the onscreen log.

You can also run stat_tracer :: TRaceR - Statistical Analysis Only if you want to run only the statistical analysis step, following the time course analysis.

Biological Model

Primary cell cultures of mouse cerebellar granules for the study of pharmacologically-induced ataxia-telangiectasia.

ATM (Ataxia-telangiectasia mutated) is a crucial serine/threonine protein kinase activated primarily by DNA double-strand breaks and oxidative stress. As a central regulator of the DNA damage response (DDR), ATM phosphorylates substrates like p53, BRCA1, and Nbs1 to maintain genomic stability. Beyond DNA damage, ATM functions as a sensor for oxidative stress, managing reactive oxygen species (ROS) levels. Deficiencies in ATM cause Ataxia-telangiectasia, characterized by neurodegeneration.

Experimental Conditions

• DMSO: control condition (KU vehicle - always present)
• UV: inducer of ROS production (always present)
• KU: ATM protein inhibitor (-> ataxia-telangiectasia mimicker)
• NAC: N-acetylcysteine (antioxidant agent)
• OMO: activator of NRF2 (a ROS-protective gene expression program)
• CBX: carbenoxolone (gap junction and pannexin-1 blocker)
• Gd: gadolinium (calcium channel blocker)
• TROLOX: Analogue of Vitamin E (antioxidant agent)

1. Batch KU_lessUV_NAC
   • DMSO
   • DMSO_NAC
   • KU
   • KU_lessUV
   • KU_NAC
2. Batch KU_OMO
   • DMSO_OMO
   • KU
   • KU_OMO_24h
   • KU_OMO_48h
3. Batch KU_Gd_CBX
   • KU
   • KU_CBX
   • KU_Gd
   • KU_Gd_CBX
4. Batch KU_TROLOX
   • KU
   • KU_TROLOX

Data Format

The input data must be organized as follows: for each batch (KU_OMO, KU_lessUV_NAC, etc.), there must be a folder containing a subfolder for each condition (e.g., DMSO, KU, KU_NAC, etc.). Since the names of these folders will be used to identify batches, groups, and perform comparisons, avoid using spaces in the folder names.

It is assumed that each input file—representing a biological replicate—consists of a CSV organized as follows:

• , for decimal point
• ; as field separator
• a string in the first row to skip
• column headings in the second row
• fileEncoding UTF-16LE (with BOM)

Finally, each experiment is expected to last 900 seconds, at a sampling rate of 1 second. Otherwise, the statistics may no longer be comparable.

Statistical Contrasts

Contrasts need to be defined inside the input folder of each batch through a plain text file, named comp.txt (or comparisons.txt), where each statistical comparison is listed in the form

condition -- reference

Blank lines allowed, as well as comment lines introduced by #.

Requisites

R packages

• r4tcpl
• stats
• dplyr
• tidyr
• ggplot2
• patchwork
• zoo
• pracma

Kerblam! (>= 1.2.1)

Anomalies

Recording Length

The standard duration of each experiment should be exactly 900 seconds, at a sampling rate of 1 second. The following files are exceptions. However, given the experimental conditions, we believe they do not introduce statistical artifacts.

Only 600 time samples for ./data/in/KU_lessUV_NAC/DMSO experiments

• 250908_011_DMSO.csv
• 250908_012_DMSO.csv
• 250908_013_DMSO.csv

Only 700 time samples for ./data/in/KU_lessUV_NAC/KU_NAC experiments

• 250908_045_KU_NAC.csv

CSV separators

./data/in/KU_OMO/KU/251113_011_KU.csv was the only CSV input file with a different separator convention, even though it used the same file encoding, UTF-16LE (with BOM). This caused the analysis in the ./data/in/KU_OMO/KU/ folder to fail, although it did not stop the entire program from running. It was fixed by:

fpath <- "./data/in/KU_OMO/KU/251113_011_KU.csv"
fpath |> read.table(
        header = TRUE,
        sep = "\t",
        dec = ".",
        skip = 1,
        check.names = FALSE,
        stringsAsFactors = FALSE,
        fileEncoding = "UTF-16LE") |>
    lapply(as_num) |> as.data.frame() |> write.table(
        file = "./data/in/KU_OMO/KU/251113_011_KU_new.csv",
        sep = ";",
        dec = ",",
        row.names = FALSE,
        col.names = TRUE,
        quote = FALSE,
        fileEncoding = "UTF-16LE")

Then the first-line string Channel.002\r\n was manually reinserted in ./data/in/KU_OMO/KU/251113_011_KU_new.csv and saved as ./data/in/KU_OMO/KU/251113_011_KU.csv.