The article introduces TranslatorX, a web server designed to align protein-coding nucleotide sequences based on their corresponding amino acid translations. The authors highlight the limitations of aligning nucleotide sequences directly, such as the rapid degradation of sequence similarity at the DNA level compared to the amino acid level. TranslatorX addresses these issues by using amino acid translations to guide the alignment process, offering several innovations including the use of all documented genetic codes, multiple alignment programs, ambiguous codon translation, and an innovative alignment cleaning criterion based on GBlocks. The server provides rich output, including graphical visualization, codon-based alignments, and measures of compositional bias. The authors demonstrate the benefits of TranslatorX through a comparative analysis of phylogenetic inference performance, showing that back-translated alignments using amino acid information result in better alignment performance and more accurate phylogenetic trees. The study also highlights the importance of considering nucleotide compositional biases and the retention of variable regions in phylogenetic analyses.The article introduces TranslatorX, a web server designed to align protein-coding nucleotide sequences based on their corresponding amino acid translations. The authors highlight the limitations of aligning nucleotide sequences directly, such as the rapid degradation of sequence similarity at the DNA level compared to the amino acid level. TranslatorX addresses these issues by using amino acid translations to guide the alignment process, offering several innovations including the use of all documented genetic codes, multiple alignment programs, ambiguous codon translation, and an innovative alignment cleaning criterion based on GBlocks. The server provides rich output, including graphical visualization, codon-based alignments, and measures of compositional bias. The authors demonstrate the benefits of TranslatorX through a comparative analysis of phylogenetic inference performance, showing that back-translated alignments using amino acid information result in better alignment performance and more accurate phylogenetic trees. The study also highlights the importance of considering nucleotide compositional biases and the retention of variable regions in phylogenetic analyses.