PaleoProPhyler: a reproducible pipeline for phylogenetic inference using ancient proteins

Abstract

Ancient proteins from fossilized or semi-fossilized remains can yield phylogenetic information at broad temporal horizons, in some cases even millions of years into the past. In recent years, peptides extracted from archaic hominins and long-extinct mega-fauna have enabled unprecedented insights into their evolutionary history. In contrast to the field of ancient DNA - where several computational methods exist to process and analyze sequencing data - few tools exist for handling ancient protein sequence data. Instead, most studies rely on loosely combined custom scripts, which makes it difficult to reproduce results or share methodologies across research groups. Here, we present PaleoProPhyler: a new fully reproducible pipeline for aligning ancient peptide data and subsequently performing phylogenetic analyses. The pipeline can not only process various forms of proteomic data, but also easily harness genetic data in different formats (CRAM, BAM, VCF) and translate it, allowing the user to create reference panels for phyloproteomic analyses. We describe the various steps of the pipeline and its many functionalities, and provide some examples of how to use it. PaleoProPhyler allows researchers with little bioinformatics experience to efficiently analyze palaeoproteomic sequences, so as to derive insights from this valuable source of evolutionary data.

Automated high-throughput biological sex identification from archaeological human dental enamel using targeted proteomics

“…There are only a few publicly available amelogenin protein sequences on Uniprot. However, efforts have been made to translate primate genomes into protein sequences 53,54 , and based on those translated sequences, we can confirm that the assay could be extended to the Catarrhini parvorder (old world monkeys) that encompasses the Cercopithecidae, the Hylobatidae and the Hominidae 55 . Assessing the sex of other species would require the study of a population large enough to take advantage of the dynamic model built on the analysed dataset.…”

PaleoProPhyler: a reproducible pipeline for phylogenetic inference using ancient proteins

“…The modules are intended to work with each other, but can also be used independently. An in depth explanation of each step of each module, as well as the code being run in the background, is provided on the software's Github page (Patramanis et al, 2023) as well as in the supplementary material.…”

“…The process of generating the reference dataset and its phyloproteomic tree using PaleoProPhyler is covered in detail in the step-by-step Github Tutorial. The dataset used as input for the creation of the phylogenetic tree is available at Zenodo (Patramanis et al, 2022) 4…”

“…Details on the creation of the protein list can be found in the supplementary materials. The palaeoproteomic hominid reference dataset is publicly available online at Zenodo, under the name 'Hominid Palaeoproteomic Reference Dataset' (Patramanis et al, 2022)…”

“…The Protein Reference Dataset is available on Zenodo: 10.5281/zenodo.7728060 (Patramanis et al, 2022) PaleoProPhyler is publicly available on Github and Zenodo: 10.5281/zenodo.10122365 (Patramanis et al, 2023) The tool requires a Linux OS (Operating System) and the installation of Conda. The github repository contains a tutorial for using the workflow presented here, with the proteins recovered from the Homo antecessor and Gigantopithecus blacki as examples.…”