WGCNA: Problem with selecting soft threshold
0
0
Entering edit mode
Fernanda • 0
@1e91af62
Last seen 5 weeks ago
Brazil
enter code here

Hi,

Hope some knows how to help me.

I am running a WGCNA with 36 samples in RNA-Seq data and I am facing some difficulties to find the exact power in my analysis. My code is:

powers <- c(c(1:10), seq(from = 12, to=20, by=2))

sft <- pickSoftThreshold(datExpr1, powerVector = powers, verbose = 5, networkType = "signed")

I obtained this results:

Power SFT.R.sq slope truncated.R.sq mean.k. median.k. max.k.

 1  0.82700 11.0000          0.859  9450.0    9550.0  11200

 2  0.71700  4.6500          0.866  5810.0    5900.0   7990

 3  0.54500  2.3600          0.953  3780.0    3820.0   5990

 4  0.24500  1.1000          0.877  2580.0    2580.0   4640

 5  0.00385  0.0874          0.756  1830.0    1800.0   3690

 6  0.09260 -0.3660          0.719  1340.0    1300.0   3010

 7  0.27800 -0.6570          0.730  1010.0     948.0   2500

 8  0.41800 -0.8350          0.755   779.0     706.0   2110

 9  0.50700 -0.9430          0.782   613.0     533.0   1810

10 0.54800 -1.0500 0.791 490.0 410.0 1570

12 0.59700 -1.2200 0.811 328.0 250.0 1230

14 0.61900 -1.3300 0.823 230.0 159.0 989

16 0.65400 -1.3900 0.847 167.0 105.0 810

18 0.70500 -1.3900 0.881 125.0 71.3 674

20 0.74500 -1.3900 0.905 95.7 49.3 568

I tested a power from 20, but I obtained a strong correlation in grey module, so I decreased for power 16 to figure out and the same trouble repeated in my analysis. Now I have doubts if I will try running WGCNA with a power of 18 or if I increase the power to 40 to see if my mean k. improves.

Hope someone can help me.

Best,

Fernanda

biologicalnetworkanalysis genomes RNA-Seq co-expression wgcna • 173 views
ADD COMMENT

Login before adding your answer.

Traffic: 352 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6