Table 4.
Codon-wise tests of positive selection on MHC class I and IIB sequences from two populations of house sparrows in northern Norway. MHC class I sequences were further divided into two groups according to their lengths: short = 233 bp, long = 237–240 bp. The model tests were performed with CODELM in the PAML 4 package. lnLb=log-likelihood value, 2ΔL=test value of likelihood ratio test (LRT), P=significance value
| N1 | PAML | Positively selected sites2 | |||||
|---|---|---|---|---|---|---|---|
| Models | lnLb | 2ΔL | P | PAML (BEB)3 | REL4 | ||
| MHC class I short | 14 | M1a (neutral) | −517.22 | Not allowed | |||
| M2a (selection) | −516.91 | 0.63 | 0.73 | - | |||
| M7 (β) | −517.29 | Not allowed | |||||
| M8 (β+ω) | −516.91 | 0.75 | 0.69 | - | - | ||
| MHC class I long | 11 | M1a (neutral) | −951.52 | Not allowed | |||
| M2a (selection) | −933.06 | 36.93 | <0.0001 | 3**, 5*, 19**, 21**, 61**, 62**, 69** | |||
| M7 (β) | −951.78 | Not allowed | |||||
| M8 (β+ω) | −932.98 | 37.61 | <0.0001 | 2*, 3**, 5*, 19***, 21**, 61***, 62***, 69*** | 3*, 19*, 21*, 61*, 62**, 69** | ||
| MHC class IIB | 24 | M1a (neutral) | −1344.49 | Not allowed | |||
| M2a (selection) | −1307.29 | 74.42 | <0.0001 | 15**, 24**, 31**, 38*, 44**, 45**, 49**, 51**, 52* | |||
| M7 (β) | −1346.38 | Not allowed | |||||
| M8 (β+ω) | −1309.58 | 73.59 | <0.0001 | 15***, 24***, 31***, 35**, 38**, 44***, 45***, 49***, 51***, 52** | 31**, 44**, 51** | ||
number of sequences, excluding sequences containing reading frame error (N= 2 MHC class I short) or stop codon (N= 1 MHC class IIB).
significance levels for selected sites *≤0.05, **≤0.01, ***≤0.001.
calculated using the Bayes Empirical Bayes (BEB) method (Yang et al. 2005).
computed in HyPhy online service (http://www.datamonkey.org/) using the REL method for site-by-site selection, the REV nucleotide substitution bias model, and the default significance level, for comparison given in the same row as PAML M8 sites.