MCP HubMCP Hub
스킬 목록으로 돌아가기

interpret-mass-spectrum

pjt222
업데이트됨 2 days ago
4 조회
17
2
17
GitHub에서 보기
기타general

정보

이 스킬은 질량 분석 데이터를 분석하여 분자식을 결정하고 단편화 패턴을 통해 구조적 특징을 식별합니다. 다양한 이온화 방법을 지원하며 동위원소 분석, 단편화 손실 및 순도 평가를 처리합니다. 알려지지 않은 화합물 해석이나 스펙트럼 데이터로부터 분자 구조 확인에 활용하세요.

빠른 설치

Claude Code

추천
기본
npx skills add pjt222/agent-almanac -a claude-code
플러그인 명령대체
/plugin add https://github.com/pjt222/agent-almanac
Git 클론대체
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/interpret-mass-spectrum

Claude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요

문서

Interpret Mass Spectrum

Read mass spectra from any common ionization method. Determine molecular ion, molecular formula, fragmentation pathways, structural features of analyte.

When Use

  • Determine molecular weight and formula of unknown compound
  • Confirm identity of synthetic product by molecular ion + fragmentation
  • Identify impurities or degradation products in sample
  • Propose structural features from characteristic fragmentation losses
  • Analyze isotope patterns to detect halogens, sulfur, metals

Inputs

  • Required: Mass spectrum data (m/z values with relative intensities, at minimum full scan spectrum)
  • Required: Ionization method used (EI, ESI, MALDI, CI, APCI, APPI)
  • Optional: High-resolution mass data (exact mass, measured vs. calculated)
  • Optional: Molecular formula from other sources (elemental analysis, NMR)
  • Optional: Tandem MS/MS data (fragmentation of selected precursor ions)
  • Optional: Chromatographic context (LC-MS or GC-MS retention time, purity)

Steps

Step 1: Identify Ionization Method and Expected Ion Types

Determine what species spectrum contains before assigning peaks:

  1. Classify ionization method:
MethodEnergyPrimary IonFragmentationTypical Use
EI (70 eV)HardM+. (radical cation)ExtensiveSmall volatile molecules, GC-MS
CISoft[M+H]+, [M+NH4]+MinimalMolecular weight confirmation
ESISoft[M+H]+, [M+Na]+, [M-H]-MinimalPolar, biomolecules, LC-MS
MALDISoft[M+H]+, [M+Na]+, [M+K]+MinimalLarge molecules, polymers, proteins
APCISoft[M+H]+, [M-H]-SomeMedium polarity, LC-MS
  1. Note polarity mode: Positive mode makes cations. Negative mode makes anions. ESI commonly uses both
  2. Check for adducts and clusters: Soft ionization often makes [M+Na]+ (M+23), [M+K]+ (M+39), [2M+H]+, [2M+Na]+ in addition to [M+H]+. Identify these before assigning molecular ion
  3. Identify multiply charged ions: In ESI, multiply charged ions at m/z = (M + nH) / n. Look for peaks separated by fractional m/z values (e.g., 0.5 Da spacing = z=2)

Got: Ionization method documented. Expected ion types listed. Adducts/clusters identified so true molecular ion can be determined.

If fail: Ionization method unknown? Examine spectrum for clues: extensive fragmentation = EI, adduct patterns = ESI, matrix peaks = MALDI. Consult instrument log if available.

Step 2: Determine Molecular Ion and Molecular Formula

Identify molecular ion peak, derive molecular formula:

  1. Locate molecular ion (M): In EI, M+. is highest m/z peak with reasonable isotope pattern (may be weak or absent for labile compounds). In soft ionization, identify [M+H]+ or [M+Na]+ and subtract adduct to get M
  2. Apply nitrogen rule: Odd molecular weight = odd number of nitrogen atoms. Even molecular weight = zero or even number of nitrogens
  3. Calculate degrees of unsaturation (DBE): DBE = (2C + 2 + N - H - X) / 2, where X = halogens. Each ring or pi bond = 1 DBE. Benzene = 4 DBE, carbonyl = 1 DBE
  4. Use high-resolution data: Exact mass available? Calculate molecular formula using mass defect. Compare measured mass with all candidate formulas within mass accuracy window (typically < 5 ppm for modern instruments)
  5. Cross-check with isotope pattern: Observed isotope pattern must match proposed molecular formula (see Step 3)

Got: Molecular ion identified. Molecular weight determined. Nitrogen rule applied. Molecular formula proposed (confirmed by HRMS if available).

If fail: No molecular ion visible in EI (common for thermally labile or highly branched compounds)? Try softer ionization. Molecular ion ambiguous? Check for loss of common small fragments from highest m/z peak (e.g., M-1, M-15, M-18 can help identify M).

Step 3: Analyze Isotope Patterns

Use isotopic signatures to detect specific elements:

  1. Monoisotopic elements: H, C, N, O, F, P, I have characteristic natural abundance patterns. For molecules containing only C, H, N, O, M+1 peak = approximately 1.1% per carbon
  2. Halogen patterns:
ElementIsotopesM : M+2 RatioVisual Pattern
35Cl / 37Cl35, 373 : 1Doublet, 2 Da apart
79Br / 81Br79, 811 : 1Equal doublet, 2 Da apart
2 Cl--9 : 6 : 1Triplet
2 Br--1 : 2 : 1Triplet
1 Cl + 1 Br--3 : 4 : 1Characteristic quartet-like
  1. Sulfur detection: 34S contributes 4.4% at M+2. M+2 peak of approximately 4-5% relative to M (after correcting for 13C2 contribution) = one sulfur atom
  2. Silicon detection: 29Si (5.1%) and 30Si (3.4%) make distinctive M+1 and M+2 contributions
  3. Compare with calculated patterns: Use proposed molecular formula to calculate theoretical isotope pattern. Overlay with observed pattern to confirm or refute formula

Got: Isotope pattern analyzed. Presence or absence of Cl, Br, S, Si determined. Pattern consistent with proposed molecular formula.

If fail: Isotope resolution insufficient (low-resolution instrument)? M+2 pattern may be unresolvable. Note limitation. Rely on exact mass and other spectroscopic data for elemental composition.

Step 4: Identify Fragmentation Losses and Key Fragment Ions

Map fragmentation pathways to extract structural info:

  1. Catalog major fragments: List all peaks above 5-10% relative intensity with m/z values
  2. Calculate neutral losses from molecular ion:
Loss (Da)Neutral LostStructural Implication
1H.Labile hydrogen
15CH3.Methyl group
17OH.Hydroxyl
18H2OAlcohol, carboxylic acid
27HCNNitrogen heterocycle, amine
28CO or C2H4Carbonyl or ethyl
29CHO. or C2H5.Aldehyde or ethyl
31OCH3. or CH2OH.Methoxy or hydroxymethyl
32CH3OHMethyl ester
35/36Cl./HClChlorinated compound
44CO2Carboxylic acid, ester
45OC2H5.Ethoxy
46NO2.Nitro compound
  1. Identify characteristic fragment ions:
m/zIonOrigin
77C6H5+Phenyl cation
91C7H7+Tropylium (benzyl rearrangement)
105C6H5CO+Benzoyl cation
43CH3CO+ or C3H7+Acetyl or propyl
57C4H9+ or C3H5O+tert-Butyl or acrolein
149Phthalate fragmentPlasticizer contaminant
  1. Map fragmentation pathways: Connect fragment ions by successive losses. Build fragmentation tree from M down to low-mass fragments
  2. Identify rearrangement ions: McLafferty rearrangement (gamma-hydrogen transfer with beta-cleavage) makes even-electron ions from carbonyl-containing compounds. Retro-Diels-Alder fragmentation characteristic of cyclohexene systems

Got: All major fragment ions assigned. Neutral losses calculated and correlated with structural features. Fragmentation tree built.

If fail: Fragments do not correspond to simple losses from molecular ion? Consider rearrangement processes. Unassigned fragments may indicate unexpected functional groups, impurities, matrix/background peaks.

Step 5: Assess Purity and Propose Structure

Evaluate overall spectrum for purity indicators. Assemble structural proposal:

  1. Purity check: In GC-MS or LC-MS, examine chromatogram for additional peaks. In direct-infusion MS, look for unexpected ions not fragments of or adducts with main analyte
  2. Background and contaminant peaks: Common contaminants: phthalate plasticizers (m/z 149, 167, 279), column bleed (siloxanes at m/z 207, 281, 355, 429 in GC-MS), solvent clusters
  3. Structural proposal: Combine molecular formula (Step 2), isotope pattern (Step 3), fragmentation (Step 4) to propose structure or set of candidate structures
  4. Rank candidates: Use fragmentation tree to rank structural candidates. Best structure explains most fragment ions with fewest ad hoc assumptions
  5. Cross-validate: Compare proposed structure with data from other techniques (NMR, IR, UV-Vis). Mass spectrum alone rarely gives unambiguous structure for novel compounds

Got: Purity assessed. Contaminants identified if present. Structural proposal (or ranked candidate list) consistent with all MS data and cross-validated where possible.

If fail: Spectrum appears to contain multiple components and chromatographic separation not used? Flag mixture. Recommend LC-MS or GC-MS reanalysis. No satisfactory structural proposal emerges? Identify which additional data (HRMS, MS/MS, NMR) would resolve ambiguity.

Checks

  • Ionization method identified. Expected ion types documented
  • Molecular ion located and distinguished from adducts, fragments, clusters
  • Nitrogen rule applied and consistent with proposed formula
  • Degrees of unsaturation calculated and accounted for in structure
  • Isotope pattern matches proposed molecular formula
  • Major fragment ions assigned with neutral losses and structural rationale
  • Fragmentation tree built from molecular ion to low-mass fragments
  • Common contaminant and background peaks identified and excluded
  • Structural proposal cross-validated with other spectroscopic data

Pitfalls

  • Misidentify molecular ion: In EI, base peak often a fragment, not molecular ion. Molecular ion = highest m/z peak with chemically reasonable isotope pattern. Adduct ions in ESI ([M+Na]+, [2M+H]+) also mistaken for molecular ion.
  • Ignore nitrogen rule: Odd-mass molecular ion requires odd number of nitrogens. Forgetting this → impossible molecular formulas.
  • Confuse isobaric losses: Loss of 28 Da could be CO or C2H4. Loss of 29 could be CHO or C2H5. High-resolution MS or additional fragmentation data needed to distinguish isobaric losses.
  • Neglect multiply charged ions: In ESI, doubly or triply charged ions at half or one-third expected m/z. Look for non-integer spacing between isotope peaks as diagnostic for multiple charges.
  • Over-interpret low-abundance peaks: Peaks below 1-2% relative intensity may be noise, isotope contributions, minor contaminants rather than meaningful fragments.
  • Assume pure sample: Many real-world spectra are mixtures. Always check chromatographic purity. Look for ions inconsistent with proposed structure.

See Also

  • interpret-nmr-spectrum — determine connectivity and hydrogen environments for structural confirmation
  • interpret-ir-spectrum — identify functional groups that explain observed fragmentation
  • interpret-uv-vis-spectrum — characterize chromophores in analyte
  • interpret-raman-spectrum — complementary vibrational analysis
  • plan-spectroscopic-analysis — select and sequence analytical techniques before data acquisition
  • interpret-chromatogram — analyze GC or LC chromatographic data coupled with MS

GitHub 저장소

pjt222/agent-almanac
경로: i18n/caveman/skills/interpret-mass-spectrum
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

연관 스킬

llamaguard

기타

LlamaGuard는 폭력 및 혐오 발언 등 6가지 안전 범주에서 LLM 입력과 출력을 조정하기 위한 Meta의 70-80억 파라미터 모델입니다. 94-95% 정확도를 제공하며 vLLM, Hugging Face 또는 Amazon SageMaker를 사용해 배포할 수 있습니다. 이 기술을 사용하여 AI 애플리케이션에 콘텐츠 필터링 및 안전 가드레일을 손쉽게 통합하세요.

스킬 보기

cost-optimization

기타

이 Claude Skill은 리소스 적정화, 태깅 전략, 지출 분석을 통해 개발자들이 클라우드 비용을 최적화할 수 있도록 지원합니다. AWS, Azure, GCP에서 클라우드 비용을 절감하고 비용 거버넌스를 구현하기 위한 프레임워크를 제공합니다. 인프라 비용을 분석하거나, 리소스를 적정화하거나, 예산 제약을 충족해야 할 때 사용하세요.

스킬 보기

quantizing-models-bitsandbytes

기타

이 스킬은 bitsandbytes를 사용하여 LLM을 8비트 또는 4비트 정밀도로 양자화하며, 최소한의 정확도 손실로 50-75%의 메모리 감소를 달성합니다. 제한된 GPU 메모리에서 더 큰 모델을 실행하거나 추론을 가속화하는 데 이상적이며, INT8, NF4, FP4와 같은 형식을 지원합니다. 이 스킬은 HuggingFace Transformers와 통합되어 QLoRA 학습 및 8비트 옵티마이저를 가능하게 합니다.

스킬 보기

dispatching-parallel-agents

기타

이 Claude Skill은 3개 이상의 독립적인 문제를 동시에 조사하고 해결하기 위해 다중 에이전트를 배치합니다. 공유 상태나 의존성 없이 해결 가능한 무관련 장애 시나리오에 맞게 설계되었습니다. 핵심 기능은 병렬 문제 해결로, 각 독립 문제 영역마다 하나의 에이전트를 할당하여 효율성을 극대화합니다.

스킬 보기