update the right path to the database, update model metrics.

README.md

@@ -14,25 +14,29 @@ task_categories:
 
 <br />
 <div align="center">
-  <h1 align="center">ExomeBench: A Benchmark for Clinical Variant Interpretation in Exome Regions 🧬</h1>
-
+  <h1 align="center">ExomeBench: A Benchmark Dataset for Clinical Variant Interpretation in Exome Regions 🧬</h1>
+  <h2 align="center">Dataset Files</h2>
   <p align="center">
-    <a href="https://www.researchsquare.com/article/rs-6115078/v1">
+    <a href="https://pubmed.ncbi.nlm.nih.gov/41283814/">
       Paper
     </a>
     |
     <a href="https://github.com/Cerebras/exome_bench">
-      GitHub
+      Evaluation Code (GitHub)
     </a>
   </p>
 </div>
 
+> ExomeBench consists of datasets and code. Datasets are licensed under Creative Commons Attribution–Non-Commercial 4.0 (CC BY NC 4.0). Code provided in ExomeBench is licensed under Apache 2.0.
+> **ExomeBench is a research benchmark. It is not a diagnostic tool and should not be used to make clinical decisions.**
+
 ## 1. Project Overview
 
-The **ExomeBench** dataset is derived from [ClinVar](https://www.ncbi.nlm.nih.gov/clinvar) [(March 2024 release)](https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/), a publicly accessible database maintained by the National Center for Biotechnology Information (NCBI). ClinVar provides comprehensive information on the clinical significance of genetic variants and their associations with human diseases. This dataset focuses on variants located in exome-specific regions and includes input sequences generated from the [Human Reference Genome (HRG, GRCh38)](https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_000001405.40/).
+The **ExomeBench** dataset is derived from [ClinVar](https://www.ncbi.nlm.nih.gov/clinvar) [(Nov 2024 release)](https://ftp.ncbi.nlm.nih.gov/pub/clinvar/tab_delimited/), a publicly accessible database maintained by the National Center for Biotechnology Information (NCBI). ClinVar provides comprehensive information on the clinical significance of genetic variants and their associations with human diseases. This dataset focuses on variants located in exome-specific regions and includes input sequences generated from the [Human Reference Genome (HRG, GRCh38)](https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_000001405.40/).
 
 This dataset provides a valuable resource for researchers and practitioners working on genetic variant analysis and its clinical implications. Exome-specific regions are critically important because they encompass all protein-coding regions of the genome, where disease-associated variants are most likely to occur. By focusing on exome-specific regions and using sequences from the Human Reference Genome, this dataset enables robust evaluation of models on clinically significant tasks.
 
+Code to fine-tune and evaluate models on this dataset using the Hugging Face Transformers library is available in [ExomeBench GitHub Repository](https://www.github.com/Cerebras/exome_bench).
 
 ## 2. Dataset Details
 
@@ -46,7 +50,7 @@ This dataset provides a valuable resource for researchers and practitioners work
 - **Exome-Specific Regions**: Filter the variants to include only those located in exome-specific regions (GENCODE v.38).
 
 ### Sequence Generation
-- **Human Reference Genome (HRG, GRCh38)**: For each variant, generate input sequences from the HRG using genome.
+- **Human Reference Genome (HRG, GRCh38)**: For each variant, generate input sequences from the HRG using the variants from the ClinVar database.
 - **Sequence Length**: The length of the sequences is a parameter, typically set to 100 base pairs (bp).
 - **Variant Positioning**: The variant is centered within the sequence, which is read in from a FASTA file.
 
@@ -134,31 +138,54 @@ ExomeBench includes **five supervised tasks**, each framed as a classification p
 <table border="1" cellspacing="0" cellpadding="8">
   <thead>
     <tr>
-      <th rowspan="2">Model</th>
-      <th colspan="5">Task<br>Acc/Precision/Recall/F1/ROCAUC</th>
+      <th align="center" rowspan="2">Model</th>
+      <th align="center" colspan="5">Task (MCC)</th>
     </tr>
     <tr>
-      <th>PV</th>
-      <th>CPS</th>
-      <th>CP</th>
-      <th>BRCA</th>
-      <th>TFG</th>
+      <th align="center">PV</th>
+      <th align="center">CPS</th>
+      <th align="center">CP</th>
+      <th align="center">BRCA</th>
+      <th align="center">TFG</th>
     </tr>
   </thead>
   <tbody>
     <tr>
-      <td>NT-Multispecies-2.5B</td>
-      <td align="center">0, 0, 0, 0, 0</td>
-      <td align="center">0, 0, 0, 0, 0</td>
-      <td align="center">0, 0, 0, 0, 0</td>
-      <td align="center">0, 0, 0, 0, 0</td>
-      <td align="center">0, 0, 0, 0, 0</td>
+      <td>STRAND<sup><a href="#citation">[1]</a></sup></td>
+      <td align="center"><strong>0.360</strong></td>
+      <td align="center"><strong>0.937</strong></td>
+      <td align="center"><strong>0.774</strong></td>
+      <td align="center"><strong>0.877</strong></td>
+      <td align="center"><strong>0.996</strong></td>
+    </tr>
+      <td><a href="https://huggingface.co/zhihan1996/DNABERT-2-117M">DNABERT-2-117M</a></td>
+      <td align="center">0.162</td>
+      <td align="center">0.876</td>
+      <td align="center">0.549</td>
+      <td align="center">0.552</td>
+      <td align="center">0.996</td>
+    </tr>
+    <tr>
+      <td><a href="https://huggingface.co/LongSafari/hyenadna-tiny-1k-seqlen-hf">HyenaDNA-Tiny-1k</a></td>
+      <td align="center">0.135</td>
+      <td align="center">0.816</td>
+      <td align="center">0.445</td>
+      <td align="center">0.700</td>
+      <td align="center">0.994</td>
     </tr>
+    <tr>
+      <td><a href="https://huggingface.co/InstaDeepAI/nucleotide-transformer-2.5b-multi-species">NT-Multispecies-2.5B</a></td>
+      <td align="center">0.306</td>
+      <td align="center">0.624</td>
+      <td align="center">0.293</td>
+      <td align="center">0.422</td>
+      <td align="center">0.991</td>
   </tbody>
 </table>
 
 </div>
 
+> **Note**: For some models and tasks, the seed settings in the [STRAND paper](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5183178) were slightly different from those used in this repository, which may lead to minor variations in the reported results. Due to this, on an overly saturated tasks like TFG, you might observe a small discrepancy in the ordering of models based on MCC values compared to those reported in the paper.
 
 ## 5. Usage
 ```python
@@ -184,11 +211,11 @@ If you find our work valuable, please consider giving the project a star and cit
 
 ```bib
 @article{ExomeBench, 
-    DOI={10.2139/ssrn.5183178}, 
-    title={Introducing STRAND: A Foundational Sequence Transformer for Range Adaptive Nucleotide Decoding}, 
-    author={Ayanian, Shant and Osborne, Andrew and Xu, Clark and Molnar, Carl and Das, Pravat and Perez, Xoab and Natalia, Vassilieva and Pondenkandath, Vinay and Kanakiya, Bhargav and Venkatesh, Ganesh and et al.}, 
+    DOI={https://doi.org/10.1093/bib/bbaf618}, 
+    title={Introducing a foundational sequence transformer for range adaptive nucleotide decoding (STRAND)}, 
+    author={Ayanian, Shant et al.}, 
     year={2025},
-    journal={Research Square Pre-print}
+    journal={Briefings in Bioinformatics, Volume 26, Issue 6}
 } 
 ```
 Thank you for your support!