Python qPCR analysis for Fragaria vesca circadian clock genes and phytohormone analysis

Corine Faehn

Full analysis of F. vesca circadian clock gene expression and phytohormone levels in differential photoperiod treatments via RT-qPCR analysis and mass spectrometry

Overview

Quantitative real-time PCR is an important technique in analyzing real-time gene expression analysis. Gene expression analysis software is expensive, and calculation of relative expression of target gene using a reference gene is relatively straight-forward. This document includes the Python script for analysis of the circadian gene expression results used in the master thesis (by Corine Faehn) as well as the updated python script used for (intended) publication, and the script used to analyse the phytohormone levels.

Background information on qPCR data

Cq values represent the number of PCR cycles needed to observe a certain threshold of fluorescence for each sample. DNA sequences are doubled each cycle, so Cq values are on a log2 scale (1 cycle = 2x original sequence abundance, 2 cycles = 4x original sequence abundance), therefore higher Cq values corresponds to lower DNA expression levels.

To normalize the Cq data, target gene Cq values can be compared to an internal control/reference gene Cq value in each sample. This gives a change in Cq value (ΔCq) for each gene, then to convert ΔCq values from a log2 to a linear scale, we take 2 to the power of -ΔCq.

the 2^-ΔCq method is appropriate where: deltaCq = (CQ gene of interest - CQ internal control).

Necessary Packages to install:

scikit_posthocs
researchpy

Content

Each run is an individual Plate file (1-62) representing one gene tested per biological replicate. Download the zipped plates file and script for full use.

Runs grouped together by biological replicate (BR).

1. Each merged BR dataframe includes:

   • clean up:
     • Drop unused columns
     • NaN values renamed
       • NTC (no template control) for samples
       • 0 for Cq values
     • Fix plate mistakes (from notes)
       • misnamed targets, well re-runs

2. Group all data in one dataframe

   • Pull out IRC's (validated later)
   • Set column names for Sample ID's

3. Check data integrity

   • Check NTC's greater than 0
     • Validate further by melt curves
   • Drop NTC's

4. Failed wells

   • Identify samples where reference gene failed (nothing to compare to, inhibits downstream analysis)
   • Drop failed wells (where Cq = 0)

5. Identify outliers using reference genes

   • IQR method
   • Samples to be dropped *Outliers + samples where ref gene failed
   • Update dataframe *Compare dataframe before/after outliers removed

6. Check reference gene stabiity

   • Plot Cq values grouped by clone and treatment
   • Statistics
     • Ordinary Least Squares (OLS)
     • Post-Hoc Tukey HSD test
       • To identify which groups are significantly different for each ref gene

7. Analyze target genes

   • Plot mean Cq values
   • Transform final dataframe

     • Calculate 2^-ΔCq

       • ΔCq = Cq (target gene) - Cq (eometric mean of reference genes)
       • Take 2^-ΔCq
   • Save new dataframe to excel
   • Summarize data

8. Plot target genes

   • Function to plot by clone groups (latitudinal origin)
   • Independent Students T-tests for hypotheses
     • Independent variable: Treatment (4 levels) Zeitgeber (13 levels)
     • Dependent variable: 2^-ΔCq