From 836456d8fe16224cd80039722ba3afc78a524040 Mon Sep 17 00:00:00 2001
From: Tim O'Donnell <timodonnell@gmail.com>
Date: Thu, 25 Jan 2018 19:34:25 -0500
Subject: [PATCH] add models_class1 download (now with mass-spec)

---
 .../models_class1/GENERATE.sh                 | 45 +++++++++++++++++
 downloads-generation/models_class1/README.md  |  9 ++++
 .../models_class1/hyperparameters.yaml        | 50 +++++++++++++++++++
 3 files changed, 104 insertions(+)
 create mode 100755 downloads-generation/models_class1/GENERATE.sh
 create mode 100644 downloads-generation/models_class1/README.md
 create mode 100644 downloads-generation/models_class1/hyperparameters.yaml

diff --git a/downloads-generation/models_class1/GENERATE.sh b/downloads-generation/models_class1/GENERATE.sh
new file mode 100755
index 00000000..a22249fc
--- /dev/null
+++ b/downloads-generation/models_class1/GENERATE.sh
@@ -0,0 +1,45 @@
+#!/bin/bash
+#
+# Train standard MHCflurry Class I models.
+# Calls mhcflurry-class1-train-allele-specific-models on curated training data
+# using the hyperparameters in "hyperparameters.yaml".
+#
+set -e
+set -x
+
+DOWNLOAD_NAME=models_class1
+SCRATCH_DIR=${TMPDIR-/tmp}/mhcflurry-downloads-generation
+SCRIPT_ABSOLUTE_PATH="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/$(basename "${BASH_SOURCE[0]}")"
+SCRIPT_DIR=$(dirname "$SCRIPT_ABSOLUTE_PATH")
+
+mkdir -p "$SCRATCH_DIR"
+rm -rf "$SCRATCH_DIR/$DOWNLOAD_NAME"
+mkdir "$SCRATCH_DIR/$DOWNLOAD_NAME"
+
+# Send stdout and stderr to a logfile included with the archive.
+exec >  >(tee -ia "$SCRATCH_DIR/$DOWNLOAD_NAME/LOG.txt")
+exec 2> >(tee -ia "$SCRATCH_DIR/$DOWNLOAD_NAME/LOG.txt" >&2)
+
+# Log some environment info
+date
+pip freeze
+git status
+
+cd $SCRATCH_DIR/$DOWNLOAD_NAME
+
+mkdir models
+
+cp $SCRIPT_DIR/hyperparameters.yaml .
+
+time mhcflurry-class1-train-allele-specific-models \
+    --data "$(mhcflurry-downloads path data_curated)/curated_training_data.with_mass_spec.csv.bz2" \
+    --hyperparameters hyperparameters.yaml \
+    --out-models-dir models \
+    --percent-rank-calibration-num-peptides-per-length 1000000 \
+    --min-measurements-per-allele 75
+
+cp $SCRIPT_ABSOLUTE_PATH .
+bzip2 LOG.txt
+tar -cjf "../${DOWNLOAD_NAME}.tar.bz2" *
+
+echo "Created archive: $SCRATCH_DIR/$DOWNLOAD_NAME.tar.bz2"
diff --git a/downloads-generation/models_class1/README.md b/downloads-generation/models_class1/README.md
new file mode 100644
index 00000000..ce784b40
--- /dev/null
+++ b/downloads-generation/models_class1/README.md
@@ -0,0 +1,9 @@
+# Class I allele-specific models (ensemble)
+
+This download contains trained MHC Class I MHCflurry models.
+
+To generate this download run:
+
+```
+./GENERATE.sh
+```
\ No newline at end of file
diff --git a/downloads-generation/models_class1/hyperparameters.yaml b/downloads-generation/models_class1/hyperparameters.yaml
new file mode 100644
index 00000000..545f8d1d
--- /dev/null
+++ b/downloads-generation/models_class1/hyperparameters.yaml
@@ -0,0 +1,50 @@
+[{
+##########################################
+# ENSEMBLE SIZE
+##########################################
+"n_models": 8,
+
+##########################################
+# OPTIMIZATION
+##########################################
+"max_epochs": 500,
+"patience": 20,
+"early_stopping": true,
+"validation_split": 0.1,
+"minibatch_size": 512,
+"loss": "custom:mse_with_inequalities",
+
+##########################################
+# RANDOM NEGATIVE PEPTIDES
+##########################################
+"random_negative_rate": 0.2,
+"random_negative_constant": 25,
+"random_negative_affinity_min": 20000.0,
+"random_negative_affinity_max": 50000.0,
+
+##########################################
+# PEPTIDE REPRESENTATION
+##########################################
+# One of "one-hot", "embedding", or "BLOSUM62".
+"peptide_amino_acid_encoding": "BLOSUM62",
+"use_embedding": false,  # maintained for backward compatability
+"embedding_output_dim": 8,  # only used if using embedding
+"kmer_size": 15,
+
+##########################################
+# NEURAL NETWORK ARCHITECTURE
+##########################################
+"locally_connected_layers": [
+    {
+        "filters": 8,
+        "activation": "tanh",
+        "kernel_size": 3
+    }
+],
+"activation": "relu",
+"output_activation": "sigmoid",
+"layer_sizes": [16],
+"dense_layer_l1_regularization": 0.001,
+"batch_normalization": false,
+"dropout_probability": 0.0,
+}]
-- 
GitLab