natsel_sitehet – a test of site heterogeneity

This app evaluates evidence for whether sites differ in their mode of natural selection (Nielsen and Yang 1998).

from cogent3 import get_app

loader = get_app("load_aligned", format="fasta", moltype="dna")
aln = loader("data/primate_brca1.fasta")

sites_differ = get_app("natsel_sitehet",
    "GNC", tree="data/primate_brca1.tree", optimise_motif_probs=False
)

result = sites_differ(aln)
result

Statistics

LR	df	pvalue
1.4048	2	0.4954

hypothesis	key	lnL	nfp	DLC	unique_Q
null	'GNC-null'	-6708.3119	24	True	True
alt	'GNC-alt'	-6707.6095	26	True	True

The models have been constructed such that site-class bins have names indicating the mode of natural selection: -ve is purifying (oomega<1); neutral (omega=1); and +ve is positive natural selection (omega>1). The two parameters of interest relating to these are the bprobs (the maximum likelihood estimate of the frequency of the site-class) and the corresponding value of omega.

result.alt.lf

GNC-alt

log-likelihood = -6707.6095

number of free parameters = 26

Global params

A>C	A>G	A>T	C>A	C>G	C>T	G>A	G>C	G>T	T>A
0.8530	3.5646	0.9734	1.6405	2.1801	6.3218	8.0814	1.2347	0.7829	1.2798

continuation

T>C
3.0292

Bin params

bin	bprobs	omega
-ve	0.1043	1.000006664710014e-06
neutral	0.8052	1.0
+ve	0.0905	19.999999970424863

Edge params

edge	parent	length
Galago	root	0.5463
HowlerMon	root	0.1364
Rhesus	edge.3	0.0649
Orangutan	edge.2	0.0235
Gorilla	edge.1	0.0075
Human	edge.0	0.0182
Chimpanzee	edge.0	0.0085
edge.0	edge.1	0.0000
edge.1	edge.2	0.0099
edge.2	edge.3	0.0364
edge.3	root	0.0233

Motif params

AAA	AAC	AAG	AAT	ACA	ACC	ACG	ACT	AGA	AGC
0.0556	0.0235	0.0344	0.0556	0.0228	0.0046	0.0008	0.0289	0.0231	0.0286

continuation

AGG	AGT	ATA	ATC	ATG	ATT	CAA	CAC	CAG	CAT
0.0140	0.0381	0.0186	0.0070	0.0128	0.0192	0.0196	0.0052	0.0238	0.0221

continuation

CCA	CCC	CCG	CCT	CGA	CGC	CGG	CGT	CTA	CTC
0.0195	0.0062	0.0006	0.0263	0.0011	0.0009	0.0023	0.0032	0.0137	0.0078

continuation

CTG	CTT	GAA	GAC	GAG	GAT	GCA	GCC	GCG	GCT
0.0125	0.0105	0.0755	0.0105	0.0303	0.0315	0.0158	0.0096	0.0014	0.0137

continuation

GGA	GGC	GGG	GGT	GTA	GTC	GTG	GTT	TAC	TAT
0.0161	0.0090	0.0067	0.0133	0.0148	0.0070	0.0069	0.0213	0.0023	0.0101

continuation

TCA	TCC	TCG	TCT	TGC	TGG	TGT	TTA	TTC	TTG
0.0221	0.0082	0.0015	0.0251	0.0018	0.0040	0.0201	0.0212	0.0078	0.0108

continuation

TTT
0.0187

Getting the individual site posterior probabilities

I’m just displaying the posterior-probabilities from the first 20 positions only.

bprobs = result.alt.lf.get_bin_probs()
bprobs[:, :20]

	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19
-ve	0.1491	0.0839	0.0000	0.1313	0.1146	0.1569	0.0843	0.1191	0.1020	0.0798	0.0760	0.1569	0.0937	0.0000	0.1563	0.5173	0.0798	0.0695	0.1146	0.1216
neutral	0.7725	0.8127	0.8643	0.7851	0.7961	0.7668	0.8125	0.7927	0.8032	0.8141	0.8164	0.7668	0.8070	0.8655	0.7670	0.4824	0.8141	0.8197	0.7961	0.7909
+ve	0.0784	0.1034	0.1357	0.0837	0.0893	0.0764	0.1032	0.0882	0.0948	0.1061	0.1076	0.0764	0.0993	0.1345	0.0766	0.0003	0.1061	0.1108	0.0893	0.0875