1. Update training case study to use v0.10.0 and simplify the doc.

2. Update all scripts to v0.10.0.
3. Add Singularity instructions in Quick Start.
4. Updated deepvariant-vcf-stats-report.md and images to v0.10.0.
5. Update trio merging. (tedyun@: can you review trio-merge-case-study.md?)

PiperOrigin-RevId: 302530370

README.md

@@ -1,6 +1,6 @@
 # DeepVariant
 
-[![release](https://img.shields.io/badge/release-v0.9.0-green?logo=github)](https://github.com/google/deepvariant/releases)
+[![release](https://img.shields.io/badge/release-v0.10.0-green?logo=github)](https://github.com/google/deepvariant/releases)
 [![announcements](https://img.shields.io/badge/announcements-blue)](https://groups.google.com/d/forum/deepvariant-announcements)
 [![blog](https://img.shields.io/badge/blog-orange)](https://goo.gl/deepvariant)
 
@@ -16,7 +16,7 @@ designed for painless integration with the
 We recommend using our Docker solution. The command will look like this:
 
 ```
-BIN_VERSION="0.9.0"
+BIN_VERSION="0.10.0"
 sudo docker run \
   -v "YOUR_INPUT_DIR":"/input" \
   -v "YOUR_OUTPUT_DIR:/output" \
@@ -30,11 +30,15 @@ sudo docker run \
   --num_shards=$(nproc) **This will use all your cores to run make_examples. Feel free to change.**
 ```
 
-For more information, see:
+If you're using GPUs, or want to use Singularity instead, see
+[Quick Start](docs/deepvariant-quick-start.md) for more details.
+
+For more information, also see:
 
-  * [Quick Start](docs/deepvariant-quick-start.md)
   * [Full documentation list](docs/README.md)
   * [Best practices for multi-sample variant calling with DeepVariant](docs/trio-merge-case-study.md)
+  * [(Advanced) Training tutorial](docs/deepvariant-training-case-study.md)
+
 
 ## How to cite
 

docs/README.md

@@ -13,7 +13,7 @@
 ## (Advanced) Training
 
 *   [Advanced Case Study: Train a customized SNP and small indel variant caller
-    for BGISEQ-500 data](deepvariant-tpu-training-case-study.md)
+    for BGISEQ-500 data](deepvariant-training-case-study.md)
 *   [DeepVariant training data](deepvariant-details-training-data.md)
 
 ## More details

docs/deepvariant-details.md

@@ -241,7 +241,7 @@ Cloud Platform. Specifying the CPU platform also allows us to report the runtime
 more consistently.
 
 ```shell
-gcloud beta compute instances create "${USER}-cpu"  \
+gcloud compute instances create "${USER}-cpu"  \
   --scopes "compute-rw,storage-full,cloud-platform" \
   --image-family "ubuntu-1604-lts" \
   --image-project "ubuntu-os-cloud" \
@@ -254,7 +254,7 @@ gcloud beta compute instances create "${USER}-cpu"  \
 ### Command for a GPU machine on Google Cloud Platform
 
 ```shell
-gcloud beta compute instances create "${USER}-gpu" \
+gcloud compute instances create "${USER}-gpu" \
   --scopes "compute-rw,storage-full,cloud-platform" \
   --maintenance-policy "TERMINATE" \
   --accelerator=type=nvidia-tesla-p100,count=1 \

docs/deepvariant-pacbio-model-case-study.md

@@ -2,20 +2,21 @@
 
 In this case study we describe applying DeepVariant to PacBio CCS reads to call
 variants. We will call small variants from a publicly available whole genome
-from PacBio.
+CCS dataset from PacBio.
 
-Starting from v0.10.0, our PacBio model is trained with additional amplified
-library data, which provides a significant accuracy boost on amplified data.
+Starting from v0.10.0, sequence from amplified libraries is included in our
+PacBio CCS training set, providing a significant accuracy boost to variant
+detection from amplified CCS data.
 
 Case study is run on a standard Google Cloud instance. There are no special
 hardware or software requirements for running this case study. For consistency
 we use Google Cloud instance with 64 cores and 128 GB of memory. This is NOT the
 fastest or cheapest configuration. For more scalable execution of DeepVariant
 see the [External Solutions] section.
 
-In v0.8 DeepVariant released a special model that works with PacBio data. In
-this case study we will apply PacBio model by specifying `PACBIO` in
-`model_type` parameter in the `run_pacbio_case_study_docker.sh` script.
+In v0.8 DeepVariant released a model for PacBio CCS data. In this case study we
+will apply PacBio model by specifying `PACBIO` in `model_type` parameter in the
+`run_pacbio_case_study_docker.sh` script.
 
 ## Case study overview
 

docs/deepvariant-quick-start.md

@@ -153,6 +153,55 @@ output.visual_report.html
 For more information about `output.visual_report.html`, see the
 [VCF stats report documentation](deepvariant-vcf-stats-report.md).
 
+## Notes on GPU image
+
+If you are using GPUs, you can pull the GPU version, and make sure to run with
+`nvidia-docker`:
+
+```
+sudo nvidia-docker run \
+  -v "${INPUT_DIR}":"/input" \
+  -v "${OUTPUT_DIR}:/output" \
+  google/deepvariant:"${BIN_VERSION}-gpu" \
+  /opt/deepvariant/bin/run_deepvariant \
+  ...
+```
+
+## Notes on Singularity
+
+### CPU version
+
+```
+# Pull the image.
+singularity pull docker://google/deepvariant:"${BIN_VERSION}"
+
+# Run DeepVariant.
+singularity run -B /usr/lib/locale/:/usr/lib/locale/ \
+  docker://google/deepvariant:"${BIN_VERSION}" \
+  /opt/deepvariant/bin/run_deepvariant \
+  --model_type=WGS \ **Replace this string with exactly one of the following [WGS,WES,PACBIO]**
+  --ref="${INPUT_DIR}"/ucsc.hg19.chr20.unittest.fasta \
+  --reads="${INPUT_DIR}"/NA12878_S1.chr20.10_10p1mb.bam \
+  --regions "chr20:10,000,000-10,010,000" \
+  --output_vcf="${OUTPUT_DIR}"/output.vcf.gz \
+  --output_gvcf="${OUTPUT_DIR}"/output.g.vcf.gz \
+  --num_shards=1 \ **How many cores the `make_examples` step uses. Change it to the number of CPU cores you have.**
+```
+
+### GPU version
+
+```
+# Pull the image.
+singularity pull docker://google/deepvariant:"${BIN_VERSION}-gpu"
+
+# Run DeepVariant.
+# Using "--nv" and "${BIN_VERSION}-gpu" is important.
+singularity run --nv -B /usr/lib/locale/:/usr/lib/locale/ \
+  docker://google/deepvariant:"${BIN_VERSION}-gpu" \
+  /opt/deepvariant/bin/run_deepvariant \
+  ...
+```
+
 ## Evaluating the results
 
 Here we use the `hap.py`
@@ -189,7 +238,7 @@ Benchmarking Summary:
 [BAM]: http://genome.sph.umich.edu/wiki/BAM
 [BWA]: https://academic.oup.com/bioinformatics/article/25/14/1754/225615/Fast-and-accurate-short-read-alignment-with
 [docker build]: https://docs.docker.com/engine/reference/commandline/build/
-[Dockerfile]: https://github.com/google/deepvariant/blob/r0.9/Dockerfile
+[Dockerfile]: https://github.com/google/deepvariant/blob/r0.10/Dockerfile
 [External Solutions]: https://github.com/google/deepvariant#external-solutions
 [FASTA]: https://en.wikipedia.org/wiki/FASTA_format
 [Quick Start in r0.7]: https://github.com/google/deepvariant/blob/r0.7/docs/deepvariant-quick-start.md