gfellerlab/mixmhcpred

HLA-I ligand predictor

mixmhcpred's Issues

Output file formatting

Is there a way to output the file to csv or other formats that are easier to work with?

Empty Output

Hello, I tried to run MixMHCpred-2.2 using linux but my output is empty. I tried using a simple list of only peptides, as well as the list with header but it still doesn't work.
test.txt

Code bugs

when I ran MixMHCpred according to the README file, the following error happens:
/ST_PRESICION/huweipeng/software/MixMHCpred-master/lib//MixMHCpred.x: 3: /ST_PRESICION/huweipeng/software/MixMHCpred-master/lib//MixMHCpred.x: Syntax error: Unterminated quoted string

./MixMHCpred.x: cannot execute binary file

Hi,

I tried to run MixMHCpred but the error is 'cannot execute binary file'. Can you let me know how can I solve the problem ?

Thanks

Interpretation of the 4th column in the output file

the header for the fourth column in the output file says %Rank_bestAllele. However, it does not match the results from latter columns.

For example, in the 33rd row of out_compare.txt file, the best allele is A0101, the 4th column is 1% but the rank for A0101 (6th column) is actually 0.3%.

May I ask why is that?

Conda package

Hi,

would it be possible to supply a conda package?

Negative scores and %rank

Hi,

I'm getting a negative score and %rank for some of my peptides, I'm not sure if this is normal?

Line 6 in PWMs

What is the meaning of the numbers in line six in the PWM files, e.g. "1.00000 2066 1" in "lib/pwm/class1_9/A0101_1.txt"? And are these numbers relevant for the software if I want to use my own PWMs?

Inquiry Regarding Dataset from Recent Publication

Hi, I recently read your team's latest publication, "MixMHCpred2.2 & PRIME2.0 (Cell Systems 2023)", and noticed that you have organized and made available a test dataset. I reviewed the contents of Table S3: List of peptides including both HLA-I ligands (i.e., positives) and random peptides (i.e., negatives) used to benchmark MixMHCpred2.2, related to Figure 2, available at https://data.mendeley.com/datasets/2kmmjp4tmm/1.

This table states that the HLA-I ligands in the first ten datasets are sourced from Gfeller et al., while those in the last eleven datasets are from Pyke et al. However, I noticed that only the names of each sub-dataset, such as 3779-AMM_I, are provided. Could you please provide information on what alleles each of these 21 datasets corresponds to?

Looking forward to your response :)

Questions about score

Does a higher score indicate a stronger affinity for the prediction, and a negative score means that the prediction result is not combined?