PANJEP::PA-based NJ with EP v1.0

PANJEP is Pairwise Alignment-based Neighbor Joining with Edge Perturbation. First, instead of using a multiple sequence alignment (MSA), the PANJEP estimates a distance matrix based on all-to-all pairwise alignments. Next, a phylogenetic tree is inferred using the Neighbor-Joining (NJ) method, and the reliability of internal branches is evaluated using the Edge Perturbation (EP) method.

Input data

PHYLIP distance matrix OR FASTA sequence file (protein/nucleotide); format and sequence type are auto-detected. (sample FASTA) (sample matrix)

OpenMP threads (-t)

Default 1 to keep concurrent web requests fair. Leave blank to use all cores.

Search sensitivity (-s)

1.0 – 7.5; FASTA input only. Protein → MMseqs2 -s; nucleotide → blastn -task (<3 megablast, 3–5 dc-megablast, 5–7 blastn, ≥7 blastn-short).

Evolutionary-distance method (-d)

FASTA only. auto = ScoreDist for protein, Poisson(nc=4) for nucleotide. scoredist/poisson/pdist work for any input; jc69/k2p/f81/f84/tn93/logdet/ry/rysym are DNA-specific (k2p/f84/tn93/logdet/ry/rysym are DNA-only).

EP iterations (-e)

FASTA only; set 0 to disable.

Print timing / statistics (-v)