Skip to content

R and Python Code used in Naqvi et al, Genome Research 2019, doi: 10.1101/gr.230433.117

Notifications You must be signed in to change notification settings

snaqvi1990/Naqvi17-code

Repository files navigation

Naqvi17-code

generate_tscan_kmers.py takes miRNA family seed sequences as input, and shuffles them, providing a specified number of shuffled sequences per miRNA family. In the manuscript, 6 shuffled 7mers per true seed were used

count_targetscan70_cons_output.py takes the output of targetscan_70.pl (available from the TargetScan website) as input, and counts gene-miRNA interactions with target sites conserved between human and a specified species

plot_transitions.py takes a basewise 3'UTR branch length file (provided as bls_big_outfile.txt) and a desired transcript, segments the transcript, and plots the output. Segmented UTRs for all 3'UTRs in bls_big_outfile.txt are provided in big_transitions.txt. Adapted from code in: https://github.com/kslin/targetscan/blob/master/conservation/bls_helpers.py

XZlogisticRegression.R reads Supplemental Tables 3 and 4 (convert to .txt first) and peforms multinomial or binomial logistic regression with or without Pct score or human pHI as a predictor of evolutionary class, using Aikake's Information Criterion (AIC) to assess model fit.

ChickenHumanLizardSexDiff.R reads in gene count and abundance files along with sample info (provided in this folder), and uses limma/voom to calculate chicken male/female ratios (normalized to ancestral human and anolis male/female ratios) in heart, liver, kidney, and brain

About

R and Python Code used in Naqvi et al, Genome Research 2019, doi: 10.1101/gr.230433.117

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published