Comments (7)
We haven't shared the sub-optimal models. I just uploaded them here: https://illumina.box.com/s/0ey4qezxeerz7kmen7w6oinqtzetcdi3 Could you let me know if you can download them? The ones with c in their names were trained using only canonical annotations, the ones with g in their names were trained using both canonical and GTEx annotations (see Methods in the paper for more details).
from spliceai.
Hi,
I could not download the files, had to register for some business account, which is not acceptable.
Could you please put the files on some FTP site or free cloud
from spliceai.
Could you try this link: https://drive.google.com/open?id=1DrnOVmyLV2rFWWTa-lbZWzP3YzwO59K2
from spliceai.
Dear Kishore,
I tried to use these models, but I got the following error when I directly passed them to spliceai (spliceai -I calls.vcf -O calls_spliceAI.vcf -R GRCh38.fa -A "models/SpliceNet10000_c1.h5"):
File "/home/user/.local/lib/python3.6/site-packages/spliceai/main.py", line 53, in main
ann = Annotator(args.R, args.A)
File "/home/user/.local/lib/python3.6/site-packages/spliceai/utils.py", line 19, in init
df = pd.read_csv(annotations, sep='\t', dtype={'CHROM': object})
and when I first passed them to the test_model.py routine(python3 -u test_model.py 1000 > Outputs/SpliceAI1000.txt, with correct input path "model[v] = load_model('models/SpliceNet' + str(CL)+ '_c' + str(version[v]) + '.h5')"), I get this error:
File "test_model.py", line 30, in
+ '_' + '0' + '.h5', 'r')
File "/home/user/.local/lib/python3.6/site-packages/h5py/_hl/files.py", line 394, in init
swmr=swmr)
File "/home/user/.local/lib/python3.6/site-packages/h5py/_hl/files.py", line 170, in make_fid
fid = h5f.open(name, flags, fapl=fapl)
File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
File "h5py/h5f.pyx", line 85, in h5py.h5f.open
OSError: Unable to open file (unable to open file: name = './dataset_test_0.h5', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0)
Could you please let me know what went wrong.
-Amir
from spliceai.
Yeah, it seems that this is hard-coded:
SpliceAI/spliceai/utils.py
Look into this.
paths = ('models/spliceai{}.h5'.format(x) for x in range(1, 6))
self.models = [load_model(resource_filename(name, x)) for x in paths]
from spliceai.
A few things are hard-coded into the software. These models are inferior compared to the ones used in the software (we trained them mainly to demonstrate the need for using a model with 10K nucleotide context and GTEx-based augmentation), keep that in mind while using these models. Apart from the lines pointed out by Christoph in the post above, there's another thing that needs to be changed in SpliceAI/spliceai/utils.py if you want to use these models:
(line 91) wid = 10000+cov
The 10000 here is a hard-coded parameter, which needs to be changed depending on the models you are interested in (for example, 2000 if you are using models with 2K nucleotide context).
from spliceai.
@amir-azizi Another mistake you're making: The -A flag is used to input the annotation file, not the models (see README). Use -A grch37 if you're working with hg19 assembly, or -A grch38 if you're working with hg38 assembly.
from spliceai.
Related Issues (20)
- Lower Accuracy Than Introme HOT 1
- Training with additional Batch Normalization layer producing strange results HOT 1
- Trouble to launch SpliceAI with grch37 HOT 5
- spliceAI not giving output value while running using vep (Variant Ensemble Predictor) HOT 3
- Position of splice sites within an insertion HOT 1
- Training input shape HOT 1
- Question about using snv and indel score files
- variant not scored HOT 5
- Running SpliceAI takes too much time
- Duplicate records in the released VCF file HOT 3
- Unable to install using conda install HOT 1
- Running Short Tandem Repeat genotypes
- build-in grch38 annotation
- How to make a custom annotation set? HOT 2
- No training configuration found in the save file, so the model was *not* compiled. Compile it manually. HOT 3
- spliceai score HOT 3
- Query about spliceai to calculate Delins HOT 1
- WARNING:root:Skipping record (ref too long)
- Way to many TEMP files
- Figure 1c Reproduction HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from spliceai.