Scripts and data accompanying Wyder et al 2017:
Consistent Reanalysis of Genome-wide Imprinting Studies in Plants Using Generalized Linear Models Increases Concordance across Datasets bioRxiv https://doi.org/10.1101/180745

• An edgeR-based analysis method to identify imprinted genes
• requires RNA-seq data for reciprocal F1 crosses
• takes into account biological replicates and count overdispersion
• outperforms commonly used count statistics (like Fisher's exact or Chi-Square tests)
• works for any tissue (including triploid endosperm)

---

Data folders

RAW_COUNTS	maternal/paternal raw counts of examined datasets
Informative_SNP_Positions	helper files for Classify_Alleles.py

---

Classify_Alleles.py

This script classifies mRNA-seq reads by strain. It takes a sorted BAM file, and a file containing positions of interest (SNPs where parental lines are homozygous for a different variant).
Requires pysam (tested with pysam v0.8.4). Installation of pysam is easiest with bioconda especially on Mac OS X (see https://github.com/pysam-developers/pysam).

Example:

python Classify_Alleles.py example_sorted.bam example_Pos_Of_Interest > Counts_Alleles_SRRxxxxxxx

Input: example_Pos_Of_Interest

A text file with six columns (tab delimited) describing position of interest annotated with overlapping GeneID. For reproducibility, this file should be sorted by chromosome and position (sort -k1,1 -k2,2nr FILE)

Chr1 37387 37388 G T AT1G01060
...

1. chromosome
2. start
3. end
4. reference allele
5. non-reference allele
6. GeneID overlapping this position

Annotation with GeneID can be done with bedtools intersect. Files ready for analysis are available in folder Informative_SNP_Positions (Arabidopsis Ler-Col and Maize B73-Mo17).

Output:

Gene MaternalReads PaternalReads AT1G06190 34 55 ... A text file with three columns

1. GeneID
2. reference read counts
3. non-reference read counts

NOTE: PCR duplicates are not excluded but can be removed before using samtools rmdup.

---

run_edgeR_LerCol_Pignatta.R

This script runs GLM analysis based on edgeR to identify statistically significantly imprinted genes. Requires RNA-seq of reciprocal F1 cross samples (at least 1 per reciprocal cross, better 2-3 samples per reciprocal cross).

Assumes that allelic count tables are located in the same directory with file names "Counts_Alleles_SRRxxxxxxx" where xxxxxxx is the SRR sample ID.

Example for ColxLer and LerxCol samples from Pignatta et al. (2014). Not fully generalized, some things are hard-coded.

Example:

Rscript run_edgeR_LerCol_Pignatta.R

NOTE: Make sure the working directory only contains count files from reciprocal crosses of 2 strains (e.g ColxLer or LerxCol). This script uses all files whose name start with 'Counts_Alleles_'

---

Authors

• Stefan Wyder