Breaking the restriction barriers and applying CRISPRi as a gene silencing tool in Pseudoclostridium thermosuccinogenes

Joyshree Ganguly¹, Maria Martin-Pascual², Diego Montiel González³, Alkan Bulut³, Bram Vermeulen², Ivo Tjalma², Athina Vidaki^3,c, Richard van Kranenburg^1,2

• 1 Corbion, Arkelsedijk 46, 4206 AC Gorinchem, The Netherlands
• 2 Laboratory of Microbiology, Wageningen University and Research, Stippeneng 4, 6708 WE Wageningen, The Netherlands
• 3 Department of Genetic Identification, Faculty of Medicine, Erasmus University Rotterdam, 3000 CB Rotterdam, The Netherlands
• 4 Fontys University of Applied Sciences, 5612 AR Eindhoven, The Netherlands
• contributed equally

Installation requirements

BLAST+ v2.6.0
python>=3.6 & anaconda3
R language >=3.6.1 "Action of the toes"

Scripts and pipieline

1. Obtain list of motifs combinations based on the substitution DNA criteria
   Usage:
   

   python get_motifs_combinations.py

   Output:
   

   Files with a list of motif combinations

   • [Included in motifs folder with extension .txt]

2. 2.1 Blast: Create a database index reference for e coli with following command:

   makeblastdb -in ecoli/HST04_SEQt.fasta -out ecoli/HST04 -dbtype nucl

   2.2 Performs a blastn search with the motifs combinations against the ecoli reference:

   blastn -query motifs/CACNNNNNNNTNGC.txt -db ecoli/HST04 -out motifs/CACNNNNNNNTNGC.out -task blastn-short -outfmt 7 -perc_identity 100 -evalue 7 -num_threads 4;

   blastn -query motifs/DGAGNNNNATC.txt -db ecoli/HST04 -out motifs/DGAGNNNNATC.out -task blastn-short -outfmt 7 -perc_identity 100 -evalue 7 -num_threads 4; blastn -query motifs/GCNANNNNNNNGTG.txt -db ecoli/HST04 -out motifs/GCNANNNNNNNGTG.out -task blastn-short -outfmt 7 -perc_identity 100 -num_threads 4;

   blastn -query motifs/TCABNNNNNNTARG.txt -db ecoli/HST04 -out motifs/TCABNNNNNNTARG.out -task blastn-short -outfmt 7 -perc_identity 100 -evalue 7 -num_threads 4;

   blastn -query motifs/CYTANNNNNNVTGA.txt -db ecoli/HST04 -out motifs/CYTANNNNNNVTGA.out -task blastn-short -outfmt 7 -perc_identity 100 -num_threads 4;

   blastn -query motifs/GATNNNNCTC.txt -db ecoli/HST04 -out motifs/GATNNNNCTC.out -task blastn-short -outfmt 7 -perc_identity 100 -evalue 7 -num_threads 4;

   Output:

   Table of hits for each motifs list (combinations) against the ecoli reference database

   • [included in motifs folder with extension .out]

3. Get A modification based on list of motifs combinations and wig files Usage:

   python get_modifications.py motifs/ 0 4

   motifs/: is the directory containing motifs combinations prodcued from get_motifs_combinations script
   mismatches [0]: number of mismatches allowed at the end of the motif
   n_threads[4]: number of cpus to use during get modifications

   • [This can be very exhaustive by looking at almost 1,000,000 (one million) positions of each wig to 200,000 motifs combinations]

   Output: list of 6mA modification from the combinations of motifs

   • [Included in modification_levels folder per barcode (wig format) for both plus and minus strand]

4. Data Analysis

   For the data analysis we use the outputs from modification_levels Here in order to compare all possible motifs and barcodes combinations we have written some functions to compare them using the R scripting language.

   For each of the comparison we cannot assume that they follow a normal distribution pattern so we decided to perform a non-parametric test with Wilcoxon also called Mann–Whitney U test (similar to a t-test where you compare mean distributions). We also do so for all the comparisons. We set a default parameter of two - sided where we do not specify which of the two distributions is greater than the other. For example the idea is to see a significant difference with barcode05 with CYTANNNNNNVTGA against barcode03 with motif CYTANNNNNNVTGA. Assuming that barcode05 did not show any 6mA for that motif.

   • [***Important set working directory where the modification levels files are] Please open the script script.R in Rstudio in case you wanna re-run the test comparisons.

   Output: comparison file with all barcodes and all motifs

   • [Included in the main directory res.out]