# Error in Step B LDA Effect Size (LEfSe)

**URL:** <https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414>\
**Category:** LEfSe\
**Created:** [April 6, 2022, 4:55pm UTC](https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414 "2022-04-06T16:55:43Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Buding0807](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/buding0807/32/1365_2.png) [@Buding0807](https://forum.biobakery.org/u/Buding0807)\
**Post date:** [April 6, 2022, 4:55pm UTC](https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414/1 "2022-04-06T16:55:44Z")

</div>

Hi,

I’m trying to proceed to step B. After searching around, I did found someone else having similar issue. The following error was received:

/galaxy\_venv/local/lib/python2.7/site-packages/rpy2/rinterface/ **init**.py:185: RRuntimeWarning: Error in La.svd(x, nu, nv) : error code 1 from Lapack routine ‘dgesdd’

warnings.warn(x, RRuntimeWarning)  
Traceback (most recent call last):  
File “/shed\_tools/testtoolshed.g2.bx.psu.edu/repos/george-weingart/lefse/a6284ef17bf3/lefse/run\_lefse.py”, line 89, in   
if params[‘rank\_tec’] == ‘lda’: lda\_res,lda\_res\_th = test\_lda\_r(cls,feats,class\_sl,params[‘n\_boots’],params[‘f\_boots’],params[‘lda\_abs\_th’],0.0000000001,params[‘nlogs’])  
File “/export/shed\_tools/testtoolshed.g2.bx.psu.edu/repos/george-weingart/lefse/a6284ef17bf3/lefse/lefse.py”, line 189, in test\_lda\_r  
z = robjects.r(‘z ← suppressWarnings(lda(as.formula(’+f+’),data=sub\_d,tol=’+str(tol\_min)+’))’)  
File “/galaxy\_venv/local/lib/python2.7/site-packages/rpy2/robjects/ **init**.py”, line 359, in **call**  
res = self.eval(p)  
File “/galaxy\_venv/local/lib/python2.7/site-packages/rpy2/robjects/functions.py”, line 178, in **call**  
return super(SignatureTranslatedFunction, self). **call** (\*args, \*\*kwargs)  
File “/galaxy\_venv/local/lib/python2.7/site-packages/rpy2/robjects/functions.py”, line 106, in **call**  
res = super(Function, self). **call** (\*new\_args, \*\*new\_kwargs)  
rpy2.rinterface.RRuntimeError: Error in La.svd(x, nu, nv) : error code 1 from Lapack routine ‘dgesdd’

Attached is the data I use.  
[bf\_ptmid\_lefse\_nosqrt campy.txt](https://forum.biobakery.org/uploads/short-url/7TMjG5GlVLyxMvdnYDvo6kvu1N6.txt) (762.5 KB)

Appreciate any help and input.

Thanks,  
Xinran

---

<div class="post-metadata">

**Author:** ![mishort](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/mishort/32/556_2.png) [@mishort](https://forum.biobakery.org/u/mishort)\
**Post date:** [April 11, 2022, 9:59pm UTC](https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414/2 "2022-04-11T21:59:00Z")

</div>

Hello Xinran,  
LEfSe will often have errors if special characters are used in labels, such as “+” or “-” like you have in your data. I recommend replacing those labels with alphanumeric ones (“positive” and “negative” for instance) and trying to re-run.  
Best,  
Meg

---

<div class="post-metadata">

**Author:** ![Buding0807](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/buding0807/32/1365_2.png) [@Buding0807](https://forum.biobakery.org/u/Buding0807)\
**Post date:** [April 19, 2022, 3:03pm UTC](https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414/3 "2022-04-19T15:03:30Z")

</div>

Hi Meg,

Thank you for helping. I tried as you suggested, but I still had the error. Then I figured that the format of cells containing relative abundance data in excel should be all scientific. Where in my data I have some cells in scientific format and others in regular format. Once I uniformed the style, the error seems to be solved.

Thanks,  
Xinran

---

<div class="post-metadata">

**Author:** ![mishort](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/mishort/32/556_2.png) [@mishort](https://forum.biobakery.org/u/mishort)\
**Post date:** [May 2, 2022, 10:39am UTC](https://forum.biobakery.org/t/error-in-step-b-lda-effect-size-lefse/3414/4 "2022-05-02T10:39:47Z")

</div>

Xinran,  
Good to hear, and thank you for sharing what fixed the problem!  
Best,  
Meg