NVIDIA BioNeMo Microservices Integrated into Claude Science for Protein Structure Prediction

NVIDIA BioNeMo Microservices Integrated into Claude Science for Protein Structure Prediction NVIDIA and Anthropic have integrated the NVIDIA BioNeMo Agent Toolkit into Claude Science, allowing AI research agents to orchestrate complex protein structure prediction pipelines across specialized microservices. The BioNeMo Agent Toolkit allows Claude Science to discover and call GPU-accelerated NIM containers for MSA Search, OpenFold3, and Boltz-2. A two-protein test system using Paracoccidioides lutzii Seh1 and proposed partner C1HCX1 showed that multiple sequence alignments are essential, as interface pTM collapsed from ~0.82–0.85 to below 0.20 without them. Both OpenFold3 and Boltz-2 independently predicted that C1HCX1 β-strands insert into Seh1's open WD40 β-propeller with core Cα-RMSD below 0.7 Å. Orchestrating Structure Prediction in Claude Science The integration connects Claude Science directly to local or cloud-hosted NVIDIA NIM microservice endpoints running on NVIDIA L40S or H100 GPUs. The BioNeMo Agent Toolkit supplies domain-specific skills that handle environment formatting and API execution. In the demonstrated three-stage pipeline, the agent retrieves target sequences from UniProt, constructs evolutionary alignments via the MSA Search NIM using UniRef30, and runs parallel single-chain and complex predictions across OpenFold3 and Boltz-2. Impact of Multiple Sequence Alignments on Prediction Quality Running comparisons without evolutionary inputs revealed that MSAs are indispensable for predicting multi-chain interactions. Without an MSA, heteromer interface pTM (iPTM) scores collapsed from 0.85 to 0.14 in OpenFold3 and from 0.82 to 0.19 in Boltz-2. Increasing the sampling budget failed to recover interface quality in the absence of MSAs. While Boltz-2 managed to fold the monomer moderately well from sequence alone, OpenFold3 monomer folding collapsed from 82 to 36 pLDDT without alignment data. Structural Alignment and Hypothesis Formation When modeled together, both independent tools converged on the same structural mechanism: the partner protein C1HCX1 does not remodel Seh1, but instead completes its open WD40 β-propeller. Superposition of monomeric and heteromeric Seh1 showed a core Cα-RMSD of 0.68 Å in OpenFold3 and 0.65 Å in Boltz-2, with C1HCX1 β-strands (residues ~305–391) inserting precisely at the WD40 velcro-closure site. While this cross-model agreement provides a strong structural hypothesis, biological binding still requires experimental verification.

通过 Claude Science 运行 NVIDIA BioNeMo NIM 微服务进行蛋白质结构预测 NVIDIA 与 Anthropic 合作将 BioNeMo Agent Toolkit 整合至科研工作台 Claude Science，使 AI 智能体能够直接调度 BioNeMo NIM 微服务，执行基于多序列比对（MSA）的蛋白质单体及复合物结构预测与跨模型比对。 BioNeMo Agent Toolkit 将生物、化学及药物研发模型封装为智能体技能，内部基准测试显示其可将任务准确率从 60% 提升至 100%，并将 Token 效率提升约一倍。 系统配置需搭载 NVIDIA L40S 或 H100 GPU 的计算环境及约 700 GB 存储空间，分别运行 msa-search、OpenFold3 和 Boltz-2 三个 NIM 微服务容器。 针对真菌 Paracoccidioides lutzii 的 Seh1 (C1GY11) 与未表征分子 C1HCX1 的预测表明，MSA 输入至关重要，缺乏 MSA 会导致复合物界面置信度（iPTM）从超过 0.82 骤降至 0.19 以下。 OpenFold3 与 Boltz-2 独立给出了高度一致的结构几何预测（核心 Cα-RMSD 分别为 0.68 Å 和 0.65 Å），显示伴侣蛋白通过特定 β-折叠股闭合了 Seh1 开放的 WD40 β-螺旋结构。 环境配置与微服务接入 在配备 L40S 或 H100 GPU 的机器上，用户可通过 Claude Science 导入 BioNeMo Agent Toolkit 技能并连接本地端点。运行该流程需部署三个 NIM 容器，包括采用 UniRef30 数据库（约 490 GB）的 msa-search，以及 OpenFold3 和 Boltz-2 结构预测微服务（共计约 30–40 GB）。 三阶段协同预测工作流 Claude Science 依次执行三阶段任务：首先通过 MSA Search NIM 为目标单链和同物种配对序列生成 A3M 比对文件；随后分别驱动 OpenFold3 和 Boltz-2 进行单体及异二聚体复合物结构预测。两种模型在配对 MSA 处理逻辑上存在差异，但均完整保留了中间文件与置信度度量以供交叉评估。 MSA 的关键作用与结构假设验证 基准对比显示，去除 MSA 后，OpenFold3 和 Boltz-2 的复合物界面 iPTM 分别从 0.85 和 0.82 跌落至 0.14 和 0.19，增加采样预算也无法弥补演化信息的缺失。结构叠合分析显示两款模型均预测 C1HCX1 的 305–391 残基段补充了 Seh1 的 WD40 魔术贴闭合边缘，为生物学实验验证提供了可复现的高置信度假说。

Agentic AI is changing how research is done. AI scientists can read papers, propose hypotheses, call models, and determine which experiments to prioritize next. First proving their value in software engineering, coding agents now write, test, and ship production code. Scientific research can be more demanding and iterative. Researchers continually evaluate evidence, refine hypotheses, and run experiments that shape subsequent decisions. Even failed experiments can lead to unexpected insights. Scientific problems also often require domain-specific tools, for example, to fold a protein or characterize a molecule. Orchestrating and operating these tools can be challenging. Similar packages may have vastly different environment requirements or APIs. A general-purpose agent might recognize that a task calls for protein folding or molecular docking, but not which model to run, how to format the request, or which parameters matter. NVIDIA BioNeMo Agent Toolkit closes that gap. The toolkit packages more than a decade of NVIDIA BioNeMo life sciences models, libraries, and workflows into agent-callable skills for biology, chemistry, genomics, and drug discovery. Built to run with any agent framework, it enables complex scientific workflows using specialized domain expertise. On internal benchmarks, BioNeMo skills raise task correctness from 60% to 100% and roughly double token efficiency. This post uses Claude Science , Anthropic’s AI workbench for scientific research, with the NVIDIA BioNeMo Agent Toolkit and NVIDIA NIM to run protein structure prediction with multiple-sequence alignment (MSA) using multiple folding models and compare results. Setup NVIDIA and Anthropic worked together to integrate the BioNeMo Agent Toolkit into Claude Science, enabling agents to discover, launch, and call BioNeMo NIM microservices directly. Claude Science runs in various configurations based on where your GPU is located and your security requirements. This tutorial focuses on running the platform on a machine with a GPU. Check the product documentation for GPU compatibility information. It’s also possible to run on a laptop and connect to a remote GPU using SSH, HPC, or cloud computing on Modal. For this tutorial, you need access to a workstation or cloud machine with an NVIDIA L40S GPU or NVIDIA H100 GPU and Claude Science installed. The machine needs about 700 GB of storage for this workflow: the msa-search NIM’s UniRef30 database is ~490 GB (using the UniRef30-only profile instead of the full 1.4 TB set), and the Boltz-2 and OpenFold3 containers are 30–40 GB total. By default, Claude Science runs in a sandbox. Running BioNeMo NIM microservices requires compute endpoints that expose local or remote GPU resources. In Claude Science, select Customize > Compute > NVIDIA BioNeMo NIM > Connect . From there, import the BioNeMo Agent Toolkit skills from GitHub, add an NVIDIA API key, and connect to local endpoints—Docker containers that use the host GPU. After importing the skills, storing the API key, and configuring the local connection, start a new project and session. Then prompt Claude to create the required NIM microservice endpoints. Download and launch the local BioNeMo NIM containers for the msa-search, openfold3 and boltz2 microservices. For msa-search, download only the UniRef30 database so it downloads only ~490 GB instead of the full 1.4 TB. Run a smoke test on each to confirm they are healthy. When prompted, select Approve endpoint for both models. Setting up these endpoints requires downloading containers for each microservice. Steps With the three NIM endpoints running, Claude Science orchestrates the structure-prediction workflow. To make the pipeline concrete, this tutorial examines the Seh1 protein and a predicted Mio-family partner from Paracoccidioides lutzii , a fungus that causes paracoccidioidomycosis. The example is motivated by Figure 4e of Han, Tsenkov, Venanzi et al., “AlphaFold Database expands to proteome-scale quaternary structures” ( bioRxiv, DOI: 10.64898/2026.03.27.714458 ). Seh1 and Mio-family proteins participate in conserved nutrient-sensing machinery, and structures from other organisms suggest that Mio can complete Seh1’s otherwise open β-propeller, making this pair a useful test of MSA-supported complex prediction. This tutorial uses a two-protein system from Paracoccidioides lutzii : the nuclear-pore protein Seh1 (C1GY11) and an uncharacterized proposed partner (C1HCX1). The example asks the structural question: How does the predicted structure of Seh1 differ when it is modeled alone versus with the proposed partner? To investigate this question, Claude Science builds evolutionary context for the target sequences using MSA. It then passes those alignments to two independent folding models, preserving the inputs and outputs so the single-chain and two-chain predictions can be examined within each model. The workflow has three stages: Generate MSAs for single-chain and species-paired sequences with the GPU-accelerated MSA Search NIM. Predict structure for a single chain and a two-chain complex with the OpenFold3 NIM. Repeat the predictions with the Boltz-2 NIM and evaluate each model independently. Figure 1. BioNeMo skills in Claude Science orchestrate MSA generation and structure prediction with MSA Search, OpenFold3, and Boltz-2 NIMs Stage 1: Generate MSAs The agent retrieves both protein sequences from UniProt and creates two types of alignment. It first creates an unpaired alignment for each protein to provide information about each chain’s structure. It then creates a paired alignment by matching related versions of the two proteins found in the same species. If changes in one protein consistently correspond to changes in the other, that pattern can help the models predict where the proteins might interact. In this run, the search returned 202 sequences for each protein. The agent recorded the source and sequence for Seh1 (384 residues) and C1HCX1 (976 residues), including checksums for verification. It didn’t trim either sequence or provide structural templates, ligands, or other constraints. Use this prompt: I’m comparing the Seh1 protein C1GY11 on its own and with C1HCX1. Load the msa-search-nim skill and download both current FASTA sequences from UniProt. Save where each sequence came from, when you downloaded it, its accession, length, checksum, and chain ID: use A for C1GY11 and B for C1HCX1. Create an A3M alignment for each protein with the MSA Search NIM. Use Uniref30_2302 only, the ColabFold search type, an E-value of 0.0001, and up to 500 sequences. Then create a paired A3M for the two proteins, in A-then-B order, using greedy pairing. Save the request, response, A3M files, endpoint, metrics, and any errors. If you cannot make the paired alignment, stop and tell me; do not quietly replace it with a different kind of alignment. Stage 2: Predict structure with OpenFold3 OpenFold3 takes a molecule list with per-chain msa (unpaired) and, for complexes, paired_msa . The agent runs two conditions with the same Seh1 sequence: the monomer (chain A with its MSA) and the heteromer (chains A and B with their MSAs plus the paired MSA). Use mmCIF output, no templates, and store every returned sample and the confidence fields the service actually exposes. Use this prompt: Load the openfold3-nim skill. Use the saved alignments to make two OpenFold3 predictions: 1. C1GY11 by itself as chain A. 2. C1GY11 as chain A with C1HCX1 as chain B. Use the paired alignment for the two-protein prediction. Return mmCIF files. Do not use templates, ligands, pockets, or other constraints. Save the request, response, structures, runtime details, and every confidence score that the service returns. These scores show model confidence only; do not say they prove the proteins interact. If paired MSA input is unavailable, save the error and stop instead of continuing with an unpaired prediction. Condition confidence_score complex_plddt_score complex_pde_score ptm_score iptm_score Seh1 monomer 0.25 82.3 0.58 0.82 — (0) Seh1 + C1HCX1 0.87 78.2 1.07 0.79 0.86 Table 1. OpenFold3 confidence scores for Seh1 monomer and Seh1–C1HCX1 complex predictions using MSA input, mmCIF output, and no templates. Scores are reported as returned by the NIM The response exposed confidence_score, complex_plddt_score, complex_pde_score, ptm_score, iptm_score (plus format, name, source ); runtime_metrics was present but empty. The monomer’s iptm_score is 0 by construction, and OpenFold3’s composite confidence_score weights the interface heavily — which is why a well-folded monomer (pTM 0.82, pLDDT 82) still scores low on the composite. Read confidence_score within OpenFold3, not across models. Stage 3: Repeat the predictions with Boltz-2 The same two conditions run through Boltz-2 with the same chain IDs and no-template policy. One architectural difference matters: Boltz-2 has no separate paired_msa field. It takes one MSA per chain and pairs internally (optional top-level concatenate_msas ), so each Boltz-2 chain received its per-chain A3M. Use this prompt: Load the boltz2-nim skill and repeat the same two predictions with Boltz-2: C1GY11 alone as chain A, then C1GY11 plus C1HCX1 as chains A and B. Use the saved per-chain alignments in the format supported by the skill. Do not turn on affinity prediction or use templates, ligands, pockets, or contact constraints. Save the request, response, structures, runtime details, PAE when available, and the confidence values returned by Boltz-2. If a value is not returned, record that instead of estimating it. Repeat the predictions without MSA using OpenFold3 and Boltz-2. Review and analyze all the results, then provide a summary. Boltz-2 generated full PAE outputs ( write_full_pae=true ). Condition confidence complex_plddt (0–1) complex_pde ptm iptm protein_iptm Seh1 monomer 0.79 0.79 1.11 0.82 — (0) — Seh1 + C1HCX1 0.77 0.75 1.33 0.76 0.82 0.82 Table 2. Boltz-2 confidence scores for Seh1 monomer and Seh1–C1HCX1 predictions using MSA input and no templates. Scores are reported as returned by the NIM; pLDDT uses a 0–1 scale Boltz-2 returned a richer confidence set — confidence_scores, ptm_scores, iptm_scores, protein_iptm_scores, complex_plddt_scores, complex_iplddt_scores, complex_pde_scores, complex_ipde_scores , plus per-chain and pairwise arrays and full pae / pde matrices (up to 1360×1360 for the complex). The one requested field the service left empty was runtime metadata ( metrics: {} ). The MSA is the load-bearing input Because the workflow also ran single-sequence (no-MSA) predictions, the value of Stage 1 can be measured directly. For the heteromer, the interface pTM (iPTM), the metric that reports on the predicted contact between the two chains, collapses without an MSA in both models: Figure 2. Heteromer iPTM with and without MSA input for OpenFold3 and Boltz-2 across five samples per condition. Both models show high interface confidence with MSA input and low interface confidence without it With MSA input, iPTM reaches 0.85 for OpenFold3 and 0.82 for Boltz-2. Without MSA input, it falls to 0.14 and 0.19, respectively. The five samples cluster tightly (SD ≤ 0.006), indicating that the difference is consistent across samples. Heteromer iPTM with MSA no MSA drop OpenFold3 0.85 0.14 0.72 Boltz-2 0.82 0.19 0.63 Table 3. Heteromer iPTM with and without MSA input for OpenFold3 and Boltz-2 Two robustness checks support this result. First, running each condition with a more generous sampling budget (OpenFold3 diffusion_samples=5 ; Boltz-2 five samples with six recycling steps and 200 sampling steps), the picture remains unchanged. The with-MSA interface remained high, and the no-MSA interface remained collapsed (identical to the single-sample values within 0.01 iPTM). Extra sampling does not substitute for evolutionary input. Second, the two model families (with different architectures and, for Boltz-2, a different MSA-pairing mechanism) land within 0.03 iPTM of each other when given the alignment, which is reassuring cross-model agreement. The monomer results show the kind of MSA dependence each model has. OpenFold3 needs the alignment even to fold the single chain (pLDDT 82 → 36 without it); Boltz-2 folds the monomer fairly well from sequence alone (0.79 → 0.73) but still cannot place the interface without an MSA. In either case, Stage 1 represents more than just an optional pre-processing step—it serves as the very foundation of the interface hypothesis. Inspect the structure the workflow produced The agent then examines whether the proposed partner changes the predicted Seh1 structure. For each model, it superposes the Seh1 chain from the monomer and heteromer and examines the WD40 β-propeller region. [ Figure 3. Predicted Seh1 WD40 β-propeller structures from OpenFold3 (a, b) and Boltz-2 (c, d), shown as monomers (a, c) and Seh1–C1HCX1 heteromers (b, d). Seh1 is blue, C1HCX1 is transparent wheat, and the C1HCX1 β-strands interacting with Seh1 are red. These are predicted, not experimental, structures The monomer and heteromer views use the same camera position after alignment of the Seh1 cores, showing where the partner strands occupy the open propeller edge. The structural analysis reproduces the source paper’s Figure 4e observation in both independent models: The Seh1 fold is completed, not remodeled. Superposing Seh1 between monomer and heteromer gives a core Cα-RMSD of 0.68 Å (OpenFold3, 313/384 residues) and 0.65 Å (Boltz-2, 307/384). The partner adds to the existing propeller; the large full-chain RMSD comes only from the ~70 residues around the blade that shift as it closes. Both models place the same partner strands. Independently, OpenFold3 and Boltz-2 position a cluster of C1HCX1 β-strands from roughly residues 305–391 against Seh1 — consistent with the paper’s “three β-strands.” The insertion is at the WD40 velcro-closure position . The partner contacts Seh1’s N-terminal strand (residues 1–29) and C-terminal strands (~331–384), the edge where a WD40 propeller closes its final blade. It’s important to note that these models predict a C1GY11–C1HCX1 interaction. However, it’s not yet verified experimentally as a true biological association. In this case, two independent models converged on the same local geometry, creating a compelling hypothesis rather than proof of binding. The real value of this workflow lies in providing a reproducible, evidence-backed way to form and inspect these structural hypotheses—leaving the final word to experimental validation. Get started BioNeMo Agent Toolkit enables Claude Science agents to generate MSAs, run NIM-based structure predictions, and preserve artifacts for review. Explore the toolkit on GitHub to apply this workflow to other monomers and complexes.