Should genes with missing data be excluded from phylogenetic analyses?

Area of Science:

• Evolutionary Biology
• Bioinformatics
• Genomics

Background:

• Phylogenetic studies aim to maximize gene inclusion while minimizing missing data.
• The impact of including genes with missing data on phylogenetic accuracy, especially in likelihood analyses, is not well understood.

Purpose of the Study:

• To investigate the costs and benefits of including genes with missing data in phylogenetic analyses.
• To determine if excluding genes with missing data is a more conservative approach.
• To assess the influence of missing data on branch length estimates.

Main Methods:

• Utilized two empirical datasets (yeast and plants) with known phylogenies.
• Introduced varying amounts of missing data into different genes.
• Tested the impact of including versus excluding incomplete genes on phylogenetic accuracy.

Main Results:

• Adding incomplete genes generally increased phylogenetic analysis accuracy compared to excluding them.
• Incomplete genes were particularly beneficial for resolving poorly supported phylogenetic nodes.
• Excluding genes with missing data often led to greater accuracy decreases than including them offered increases.
• No consistent bias in branch length estimates due to missing data was observed.

Conclusions:

• Including genes with missing data is often beneficial for phylogenetic accuracy.
• Excluding genes with missing data is not necessarily a safer or more conservative approach.
• Missing data did not consistently bias branch length estimates in this study.