pXg v2.4.4

Updates

• Remove the option to output full sequences of unmapped reads to a separate file, as it consumes excessive memory and is rarely needed in typical applications.
• Update the PTM parser for compatibility with Casanovo version 5.0.0.
• Add an option to skip comment lines.
• Change the delta score calculation to a normalized delta score.
• Add an option to skip PSMs below a minimum score threshold (default: 0).

Full Changelog: v2.4.4-2025-08-11...v2.4.4-2025-08-11

pXg v2.4.3

Updates

• Introduce normalized delta score (nDeltaScore) to replace the previous delta score metric.
• Remove the original delta score.
• Add an option to specify a minimum read count or RPHM threshold (default: 1). Only peptides meeting or exceeding this threshold will be reported.
• Change the default flank size to 10.
• Implement a basic BAM indexer for cases without a .bai file. Note: this does not support multi-threading, so we recommend using samtools for indexing.
• Remove MeanMQScore from the generated PIN file.
• No longer support DB generation from pXg result unless we find a better way to utilize it.

Full Changelog: v2.4.2-2025-05-27...v2.4.3-2025-07-21

pXg v2.4.2

Updates

• Rename the field IsCanonical to IsReference.
• Add a one-hot encoded IsReference value to the PIN file.
• Implement functionality to account for amino acid variants.

Full Changelog: v2.4.1a-2025-05-01...v2.4.2-2025-05-27

pXg v2.4.1a

Updates

• Fix that some coding sequences are reported as frameshift.
• Fix a parameter confliction between pXg and BuildSequenceDB.
  • It results in omitting decoy records when dealing with multiple bam files.
• Fix strandedness typo.
• Can parse more various PTM patterns in the sequence.

Full Changelog: v2.4.1-2025-03-13...v2.4.1a-2025-05-01

pXg v2.4.1

Updates

• Added an option to select raw read count or normalized read count (i.e., RPHM, read-per-hundread-million). RPHM is the preferred choice when analyzing multiple BAM files, and it is set as the default.
• Added an option to print target-only results, which is useful for assigning read counts to confident peptides (e.g., those already verified by database search).
• Implemented a new function to build a database from FASTA files and pXg results. This functionality is available in "BuildSequenceDB.jar." The potential applications of this database are still being explored.

Full Changelog: v2.4.0-2024-12-10...v2.4.1-2025-03-13

pXg v2.4.0

Updates

• Some of parameter names were changed.
• "mode" parameter was changed to indicate strandedness.
  • Six-frame translation is used only for "none-strandedness."
  • In the paired-ends, it automatically determines the strandedness.
  • In the single-end, a user must specify a strandedness.
• Several columns were changed.
  • "InferredPeptide" presents not only the translated sequence but also the PTM annotation.
  • Length feature was introduced in PIN format
  • "MainScore" was changed to "Score" in PIN format

TODO

• Automatic determination of strandedness in a single-end dataset

Full Changelog: v2.3.2-2024-09-09...v2.4.0-2024-12-10

pXg v2.3.2

• Deal with "NULL" transcript or gene information in GTF file, particularly from Ensembl.

Full Changelog: v2.3.1-2024-07-29...v2.3.2-2024-09-09

pXg v2.3.1

• Report flank sequences, which are calculated by a frequency-based method.
• Do not report full RNA-seq sequences (unknown.seq) for unknown event of decoy matches.
• Do not write SAM file for decoy matches.
• Remove a debugging-purpose annotation in transcriptIDs column.

Full Changelog: v2.3.0-2024-05-15...v2.3.1-2024-07-29

pXg v2.3.0

• Now it can deal with multiple SAM/BAM files as input. Read count for each SAM/BAM file is written at the end of the column.
  The sum of reads from multiple SAM/BAM files is written in "Reads" column.
• To reduce memory usage, matched nucleotide sequences are indexed.

Full Changelog: https://github.com/HanyangBISLab/pXg/commits/v2.3.0-2024-05-15