Welcome to the LOng-read General github repo!

This is a GitHub repo of (mostly independent) Python/R scripts that I developed to analyse data from long-read sequencing experiments. Purpose of scripts vary from generating txt files to run community tools (example pipelines), generating plots post-SQANTI, running differential expression analyses to more custom applications.

Processing ONT raw data

A pipeline for processing raw ONT reads from transcriptome cDNA processing, using research community tools (i.e. Porechop,Minimap2,SQANTI3) and own custom scripts.

Data exploration post-SQANTI

Below listed are features that can be explored on <sample>_classification.txt generated from SQANTI.

• number of isoforms by structural category
• correlate exon number, gene length with isoform number
• identify long-non-coding RNA isoforms
• plot and test the number of isoforms with/without certain features (i.e. within/without 50bp of CAGE peak/TSS/TTS)

To run functions, read in <sample>_classification.txt file using:

• SQANTI_class_preparation(<sample>_classification.txt, standard) if expression columns are included in the file (after running --FL_count in SQANTI)
• SQANTI_class_preparation(<sample>_classification.txt, nstandard) if expression is not included

Characterize merged datasets

• subset_targetgenes_classfiles.py: Subset SQANTI classification file based on genes and reads
• colour_transcripts_by_countandpotential.py: Colour bed file by abundance and coding potential
• extract_fasta_bestorf.py: Create a fasta file based on best ORF defined from CPAT

Differential expression analysis

Current script dump to maintain. Scripts to input results after running tappAS, running linear regression etc...

Miscellaneous

• replace_filenames_with_csv.py: Replace multiple file names in a directory using reference csv file
• search_fasta_by_sequence.py: Subset fasta based on sequence
• subset_fasta_gtf.py: Subset gtf, fasta and bed files based on list of transcript IDs