The Ramp Atlas

The Ramp Atlas is an online repository of tissue and cell type-specific ramp sequences. This repository is intended to make publicly available all code and accompanying scripts that were used to retrieve, organize, and analyze data for the Ramp Atlas.

All datasets downloaded from the Human Protein Atlas were modified into Comma Seperated Values files (CSVs).
Each dataset was formatted as follows: "Gene accession, Gene Name, Tissue, na, na, Normalized Expression"
After processeing, the final files also included "Ramp Presence, Expression Bin"
These files will be reffered to as "Ramp Presence CSV" in the following input/output explanations.

Extract Ramps Directory

highGenes.py
input: original dataset CSV
output: files containing all genes in the top expression quartile for every tissue/cell-type useful for running tissue-specific ExtRamp

rampOrNot.py input: original dataset CSV
output: original CSV with added "Ramp Presence" column

covidRampOrNot.py
input: FASTA file of covid genes and sequences
output: CSV of "Covid gene, tissue, Ramp Presence"

splitDoubleLines.py
input: output of rampOrNot.py
output: genes represented by multiple isoforms in a single tissue/cell-type are seperated one line per isoform

addExpressionBins.py
input: output CSV file from splitDoubleLines.py; name for new output file
output: original CSV file with added Ramp Presence and Expression Bin columns. Expression bins are based on quartile of expression, labeled Not detected, Low, Medium, and High

removeConflictingIsoforms.py
input: output of addExpressionBins.py
output: the input CSV with all isoforms with conflicting ramp presence for the same gene removed. This is the final Ramp Presence CSV

Data Summary Directory

geneAndRampTotal.py
input: Ramp Presence CSV
output: total counts of genes and ramps per tissue and cell type, useful for dataset-wide chi-squared tests and pairwise z-tests

combinedDatasetSummary.py
input: cosnensus Ramp Presence CSV; FANTOM5 Ramp CSV; GTEx Ramp CSV; HPA Ramp CSV; name of output file
output: CSV of "Gene, Genename, Tissue, NX, Expression bin, Cosnensus ramp presence, FANTOM5 Ramp, GTEx Ramp, HPA Ramp, Total support %, Total matches, Total files" where total files is the number of datasets including that gene in that tissue, total matches is the number of datasets with ramp presence consistent with the consensus dataset, and Total support % = Total matches / Total files

rampCountByExpressionAndTissues.py
input: Ramp Presence CSV
output: counts of ramps across expression bins (Not detected, low, med, high) for each tissue/cell-type

rampCountByExpressionAndGenes.py
input: Ramp Presence CSV
output: counts of ramps across expression bins (Not detected, low, med, high) for each gene

Z-Test Directory

pairwiseZTest.py
input: output of geneAndRampTotal.py
output: CSV of "Tissue/Cell-type 1, Tissue/Cell-type 2, z-statistic, p-value"

ZTestSummary.py
input: output of geneAndRampTotal.py; number of tissue/cell-type pairs (number of tissues-1 FANTOM5=44, GTEx=33, HPA=42, consensus=61, cell=65); output of pairwiseZtest.py
output: CSV of "Tissue, Total genes, Total ramps, Significant pairs, Percent significant pairs"