How to group Italian samples in order to obtain good references?

[gixajo]

Target: gixajo_scaled
Distance: 3.4995% / 0.03499517
76.2 NorthEast_Italian
13.8 Sardinian
10.0 NorthWest_Italian

Target: gixajo_dad_scaled
Distance: 2.9711% / 0.02971138
67.2 NorthWest_Italian
22.6 Sardinian
10.2 NorthEast_Italian

Target: gixajo_mom_scaled
Distance: 4.5184% / 0.04518395
66.0 NorthEast_Italian
26.0 NorthWest_Italian
8.0 Sardinian

Dentro de poco podremos mezclar todas las medias ibéricas, italianas, británicas y francesas e ir ampliando el círculo.

[gixajo]

hay una media que aún siendo pequeña la distancia me parece un poco rara, debería tener más ceros en su distancia, voy a volverla a hacer....

[Luso]

Spoiler!

South_Italian,0.10736598,0.14725883,-0.00537531,-0.04065114,0.01536774,-0.01480103,-0.00041355,-0.00131084,0.00544821,0.02098400,0.00147906,0.00318490,-0.00570559,0.00030105,-0.00697726,-0.00277458,0.00198963,-0.00014346,0.00262584,-0.00482755,-0.00234661,0.00195073,0.00064253,0.00137365,0.00009691
NorthEast_Italian,0.12542900,0.14592190,0.03352107,-0.00264457,0.03564947,-0.00060257,0.00285707,0.00234943,0.00940747,0.02095333,-0.00253820,0.00509107,-0.01078937,-0.00234117,-0.00259897,-0.00018743,0.00269963,0.00150253,0.00365313,-0.00328243,-0.00198453,0.00203360,-0.00081240,0.00217507,-0.00182223
Central_Italian,0.11701408,0.14925148,0.00917310,-0.02814708,0.02221933,-0.01003865,0.00044550,-0.00159775,0.00642820,0.02364700,0.00002603,0.00452040,-0.00955638,-0.00467495,-0.00375115,0.00052440,0.00259400,0.00045623,0.00235623,-0.00327855,-0.00208228,0.00226180,-0.00011475,0.00257140,-0.00120693
NorthWest_Italian,0.12070933,0.14655777,0.02910123,-0.00726750,0.03182627,-0.00322107,0.00253410,0.00137310,0.00910150,0.02380610,-0.00491210,0.00471067,-0.01256423,-0.00458970,0.00161273,-0.00296987,-0.00011317,0.00007823,0.00184990,-0.00120490,0.00003520,0.00159713,0.00020753,0.00423140,-0.00031917
Sardinian,0.1216872,0.1672855,0.0284897,-0.0506523,0.060151,-0.0221337,-0.0039523,0.0024964,0.0415743,0.077351,-5.9e-05,0.0166487,-0.0286645,-0.012974,-0.013572,-0.0030013,0.0114027,-0.0013475,0.0016798,-0.0129947,-0.0021212,-0.0011018,-0.010084,-0.021383,0.0003375

I think it´s done.

Target: Josef_scaled
Distance: 4.3584% / 0.04358366
89.8 NorthWest_Italian
10.2 Sardinian

Distance to: Josef_scaled
0.04468224 NorthWest_Italian
0.04924307 NorthEast_Italian
0.06295546 Central_Italian
0.07749564 South_Italian
0.09653224 Sardinian

[gixajo]

Target: Josef_scaled
Distance: 4.3584% / 0.04358366
89.8 NorthWest_Italian
10.2 Sardinian

I changed NorthWest_Italian coordinates in the sources:

NorthWest_Italian,0.12070933,0.14655777,0.02910123,-0.00726750,0.03182627,-0.00322107,0.00253410,0.00137310,0.00910150,0.02380610,-0.00491210,0.00471067,-0.01256423,-0.00458970,0.00161273,-0.00296987,-0.00011317,0.00007823,0.00310690,-0.00120490,0.00003520,0.00159713,0.00020753,0.00423140,-0.00031917

[Luso]

I changed NorthWest_Italian coordinates in the sources:

NorthWest_Italian,0.12070933,0.14655777,0.02910123,-0.00726750,0.03182627,-0.00322107,0.00253410,0.00137310,0.00910150,0.02380610,-0.00491210,0.00471067,-0.01256423,-0.00458970,0.00161273,-0.00296987,-0.00011317,0.00007823,0.00310690,-0.00120490,0.00003520,0.00159713,0.00020753,0.00423140,-0.00031917

that source didn't change my results

[gixajo]

Hombre, sirve para probar tu teoría de que escogiendo samples random de cada país cercano podría modelar a todos los españoles, está claro que no. Obviamente una muestra italiana del norte + sur de Francia se podría modelar bastante bien a un español.

Es que yo no dije eso, yo no dije coger samples random, yo dije modelar con todas las referencias de francia e italia, y añadir dos samples sueltos que nos rasquen un poco mas de distancia.

[gixajo]

that source didn't change my results

I know but it´s more accurate.

[gixajo]

Porqué sale el Valle de Aosta NorthEast? Tiene mucho alemán por estar pegado a Suiza?

[Rocinante]

Porqué sale el Valle de Aosta NorthEast? Tiene mucho alemán por estar pegado a Suiza?

El valle de Aosta no es porque sea aleman, es porque es extremadamente frances. Su genetica no se diferencia en nada a los suizos franceses y por eso estan muy northwestern shifted, por eso yo tu no lo hubiese metido en el average.

[gixajo]

El valle de Aosta no es porque sea aleman, es porque es extremadamente frances. Su genetica no se diferencia en nada a los suizos franceses y por eso estan muy northwestern shifted, por eso yo tu no lo hubiese metido en el average.

Bueno, esto no es definitivo, se puede cambiar, hasta que no haga el thread correspondiente, este thread es para hacer pruebas.