Welcome to the pymportx documentation site!

pymportx is a Python package designed for fast gene count estimation using transcript quantification files generated by pseudoalignment or quasi-mapping tools. It is a Python adaptation of the widely-used tximport R package from Bioconductor.

Based on the tximport R package, it offers the equivalent benefits, including:

• Adjusts for potential variations in gene length between samples.
• Its upstream quantification methods (Salmon, Sailfish, Kallisto, and RSEM) are significantly faster and computationally lighter compared to alignment-based methods that require storage of BAM files.
• Increased sensitivity is achieved by retaining fragments capable of aligning to multiple genes with homologous sequences.

Prerequisites

---

Before installing pymportx, ensure you have the following prerequisites:

• Python 3.6 or later: You can download it from the official Python website.
• pip: The Python package installer. You can install pip by following the instructions here.

Installation

---

You can install pymportx via pip, the recommended Python package manager. Follow the steps below to install the package:

1. Open a terminal or command prompt

Open your terminal (Linux/macOS) or command prompt (Windows).

2. Run the following command

Install pymportx by executing the following pip command:

pip install pymportx

Data Preparation

---

We will start by creating a list of paths for each sample contained in the output directory of the upstream quantification method used:

folders = ["quant_FilePath_Sample1", "quant_FilePath_Sample2", ...]
#OR
folders = list("quant_FilePath_Sample1", "quant_FilePath_Sample2", ...)

Next, the path to a two column transcript annotation file in .csv format must be provided:

tx2gene = "path/to/tx2gene.csv"

Its first column should contain transcript IDs, while the second one must contain gene IDs, as shown below:

Usage tutorial

---

Salmon

Use the main function salmon.read_salmon() with the appropriate arguments described in the pymportx section.

from pymportx import salmon

out = salmon.read_salmon(folders,
                         tx_out=False,
                         tx2gene=tx2gene,
                         countsFromAbundance='no')

Sailfish

Use the main function sailfish.read_sailfish() with the appropriate arguments described in the pymportx section.

from pymportx import sailfish

out = sailfish.read_sailfish(folders,
                         tx_out=False,
                         tx2gene=tx2gene,
                         countsFromAbundance='no')

kallisto

Use the main function kallisto.read_kallisto() with the appropriate arguments described in the pymportx section.

from pymportx import sailfish

out = kallisto.read_kallisto(folders,
                         tx_out=False,
                         tx2gene=tx2gene,
                         countsFromAbundance='no')

RSEM

Use the main function rsem.read_rsem() with the appropriate arguments described in the pymportx section.

from pymportx import rsem

out = rsem.read_rsem(folders,
            tx_in=True
            tx_out=False,
            tx2gene=tx2gene,
            countsFromAbundance='no')

Output

---

The output of our pymportx package is an Anndata file containing the counts, abundance and legth matrices. Its structure is the following:

See the Workflow section for more details on downstream analysis using pymportx.

Additionally, Salmon, Sailfish, and kallisto users have the option to include inferential replicates (DropInfReps = False) of each sample in another DataFrame nested dictionary. The DataFrame for the inferential replicates of the first sample:

out['infReps'][0]

Or even their variance, configuring varReduce = True :

out['variance']

License

The pymportx package is released under MIT license.