Thanks for your responses. Apologies if my query was unclear. The operation I would like to do is the opposite: instead of converting to ascii, I want to convert from ascii to curv format. 

I used the command stated in your replies to do the first conversion to ascii, which I required in order to carry out an algorithm for correcting for site effects on the dataframe of vertex-wise thickness values. Now that I have done that, I was hoping to convert back into the original curv format in order to carry out the normal group comparison pipeline (mri_preproc, mri_glmfit). Is this possible, from your view? 

A slightly tangential note: when I did the first conversion to the ascii format, I used lh.inflated as the surface-- would this have had a significant impact on what I am attempting to do now?

Thanks for your help,

JeeSu

--

Date: Wed, 3 Jul 2019 18:51:49 -0400 (EDT)
From: Bruce Fischl <fischl@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Error in mris_convert thickness file from
        ascii format
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Message-ID:
        <alpine.LRH.2.20.1907031851200.2160@door.nmr.mgh.harvard.edu>
Content-Type: text/plain; charset="utf-8"

Hi JeeSu
if you look at the help for mris_convert, you'll see this example:

Convert a scalar overlay file in "curv" format to ascii:
   mris_convert -c lh.thickness lh.white lh.thickness.asc

which you should follow

cheers
Bruce

On Wed, 3 Jul 2019, 
Jee Su Suh wrote:

> 
> ????????External Email - Use Caution????????
> 
> Hello,
> 
> I am trying to convert ascii files containing vertex-wise thickness data into binary curvature files
> for subsequent smoothing and input into mri_glmfit. The command I used is:
> 
> mris_convert lh.test.asc lh.test.thickness
> 
> 
> But then get the following error:
> 
> ERROR: MRISalloc: nfaces=-5 < 0
> 
> 
> No such file or directory
> 
> 
> I'm not sure what file or directory the error message is referring to, nor what the meaning of
> nfaces is. I am running version 6.0 on macOS Mojave. I have also attached a truncated version of the
> ascii file for inspection purposes (the whole thing was too big), which was converted from a csv
> that was generated by a python script correcting for site effects. This script took in as input the
> original ascii converted from the original lh.thickness file using mris_convert. The coordinates
> correspond to the fsaverage template (~160k vertices in the whole file).
> 
> Thanks in advance for any guidance,
--