Fst - Revision history

Thorfinn: /* Note about fst for folded spectra */

2019-08-17T01:38:16Z

Note about fst for folded spectra

← Older revision		Revision as of 03:38, 17 August 2019
Line 20:		Line 20:

	=Note about fst for folded spectra=		=Note about fst for folded spectra=
			* Earlier versions of angsd/realSFS could output folded 1d sample allele frequencies which would be usefull for 1population neutrality test like Tajima. This is however not appropriate to use for calculating fst since the folding was done within population.

			* We have therefore added a proper folding procedure for the optimization based on the UNFOLDED .saf.idx files generated by -doSaf. These are the ones that should be used for calculating fst.
			Therefore please remember to add -fold 1 if you want angsd (the realSFS subfunction) to perform fst and pbs estimation using the folded spectra.

	=Two Populations real data=		=Two Populations real data=

2019-08-17T01:31:00Z

← Older revision		Revision as of 03:31, 17 August 2019
Line 18:		Line 18:
	./realSFS fst index [saf.idx's] -whichFst 1		./realSFS fst index [saf.idx's] -whichFst 1
	</pre>		</pre>

			=Note about fst for folded spectra=


	=Two Populations real data=		=Two Populations real data=
	<pre>		<pre>

2017-10-12T03:11:39Z

realSFS fst print

← Older revision		Revision as of 05:11, 12 October 2017
Line 90:		Line 90:
	;-type 0 Split out the genome into blocks. And use the first window that have data for the entire window. Then we will have the same windowcenters across datasets.		;-type 0 Split out the genome into blocks. And use the first window that have data for the entire window. Then we will have the same windowcenters across datasets.
	=realSFS fst print=		=realSFS fst print=
	You can print out the precalculated ~~alphas~~ and ~~betas~~ with		You can print out the precalculated A and B with

	''./realSFS fst print pop1.pop2.fst.idx''		''./realSFS fst print pop1.pop2.fst.idx''
Line 101:		Line 101:
	</pre>		</pre>

	The weighted fst for a region is the ratio between the sum of ~~alphas~~ and the sum of ~~betas~~. The unweighted is the mean of the persite ratios.		The weighted fst for a region is the ratio between the sum of As and the sum of B. The unweighted is the mean of the persite ratios.

			A is the alpha from the reynolds 1983 (or Bhatia) and B is the alpha + beta.

2017-09-08T07:39:00Z

realSFS fst print

← Older revision		Revision as of 09:39, 8 September 2017
Line 94:		Line 94:
	''./realSFS fst print pop1.pop2.fst.idx''		''./realSFS fst print pop1.pop2.fst.idx''

	~~Or with a nice example:~~
	Assuming we have pop1.saf.idx, pop2.saf.idx.		Assuming we have pop1.saf.idx, pop2.saf.idx.
	<pre>		<pre>
Line 101:		Line 100:
	./realSFS fst print pop1.pop2.fst.idx		./realSFS fst print pop1.pop2.fst.idx
	</pre>		</pre>

			The weighted fst for a region is the ratio between the sum of alphas and the sum of betas. The unweighted is the mean of the persite ratios.

2017-09-08T07:30:43Z

realSFS fst print

← Older revision		Revision as of 09:30, 8 September 2017
Line 95:		Line 95:

	Or with a nice example:		Or with a nice example:
	Assuming we have pop1.saf.idx, pop2.saf.idx. ~~This is per populations sample allele frequencies from two different population.~~		Assuming we have pop1.saf.idx, pop2.saf.idx.

	<pre>		<pre>
	./realSFS pop1.saf.idx pop2.saf.idx >pop1.pop2.saf.idx.ml		./realSFS pop1.saf.idx pop2.saf.idx >pop1.pop2.saf.idx.ml

2016-12-15T08:15:28Z

realSFS fst print

← Older revision		Revision as of 10:15, 15 December 2016
Line 94:		Line 94:
	''./realSFS fst print pop1.pop2.fst.idx''		''./realSFS fst print pop1.pop2.fst.idx''

	Or ~~full~~ example:		Or with a nice example:
	Assuming we have pop1.saf.idx, pop2.saf.idx~~, pop3~~.saf.idx.		Assuming we have pop1.saf.idx, pop2.saf.idx. This is per populations sample allele frequencies from two different population.
	~~<example>~~
	./realSFS fst		<pre>
	</~~example~~>		./realSFS pop1.saf.idx pop2.saf.idx >pop1.pop2.saf.idx.ml
			./realSFS fst index pop1.saf.idx pop2.saf.idx -fstout pop1.pop2 -sfs pop1.pop2.saf.idx.ml
			./realSFS fst print pop1.pop2.fst.idx
			</pre>

2016-12-15T08:04:16Z

realSFS fst print

← Older revision		Revision as of 10:04, 15 December 2016
Line 93:		Line 93:

	''./realSFS fst print pop1.pop2.fst.idx''		''./realSFS fst print pop1.pop2.fst.idx''

			Or full example:
			Assuming we have pop1.saf.idx, pop2.saf.idx, pop3.saf.idx.
			<example>
			./realSFS fst
			</example>

2016-12-15T08:02:12Z

realSFS fst print

← Older revision		Revision as of 10:02, 15 December 2016
Line 92:		Line 92:
	You can print out the precalculated alphas and betas with		You can print out the precalculated alphas and betas with

	./realSFS fst print		''./realSFS fst print pop1.pop2.fst.idx''

2016-10-07T10:48:01Z

← Older revision		Revision as of 12:48, 7 October 2016
Line 13:		Line 13:
	- Use realSFS to extract the the fst values from the '''fst'''		- Use realSFS to extract the the fst values from the '''fst'''

			NB;
			In the latest github version there is a different fst estimator which should be preferable for small sample sizes. Feel free to try that out with
			<pre>
			./realSFS fst index [saf.idx's] -whichFst 1
			</pre>
	=Two Populations real data=		=Two Populations real data=
	<pre>		<pre>

2016-01-19T20:03:36Z

realSFS fst print

← Older revision		Revision as of 22:03, 19 January 2016
Line 85:		Line 85:

	;-type 0 Split out the genome into blocks. And use the first window that have data for the entire window. Then we will have the same windowcenters across datasets.		;-type 0 Split out the genome into blocks. And use the first window that have data for the entire window. Then we will have the same windowcenters across datasets.
	==realSFS fst print==		=realSFS fst print=
			You can print out the precalculated alphas and betas with

			./realSFS fst print