Launch a Full Peptide Screening Campaign, Model Properties in Parallel¶

Define a multi-stage pipeline, run stability and solubility predictions in parallel, and rank your library down to a shortlist.

Postman API Docs

Python SDK Docs

What you'll learn:

Defining a multi-stage pipeline with parallel predictions
Filtering by melting temperature and solubility
Exploring results with summary(), stats(), and SQL queries

export BIOLMAI_TOKEN=your-token-here

Setup¶

import os
from biolmai.pipeline import (
    DataPipeline, DuckDBDataStore,
    ThresholdFilter, RankingFilter,
    ValidAminoAcidFilter, EmbeddingSpec,
    DiversitySamplingFilter,
)

TOKEN = os.environ.get("BIOLMAI_TOKEN", "")
if not TOKEN:
    raise EnvironmentError(
        "Set BIOLMAI_TOKEN before running.\n"
        "Get one at https://biolm.ai/ui/accounts/user-api-tokens/"
    )

Peptide library¶

30 antimicrobial peptides of varying length, charge, and hydrophobicity.

MY_PEPTIDES = [
    # Magainins / frog-derived
    "GIGKFLHSAKKFGKAFVGEIMNS",
    "GIGKFLHSAGKFGKAFVGEIMKS",
    "GLFDIIKKIAESF",
    "GLFDIVKKVVGALGSL",
    "FLPLILRKIVTAL",
    # Human defensins / cathelicidins
    "LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES",
    "RLFDKIRQVIRKF",
    "KWKLFKKIPKFLHLAKKF",
    # Insect-derived
    "GIGAVLKVLTTGLPALISWIKRKRQQ",
    "VDKGSYLPRPTPPRPIYNRN",
    # Synthetic / designed
    "KLAKLAKKLAKLAK",
    "LKLLKKLLKLLKKL",
    "RRWWRRWWRR",
    "KWKWKWKWKW",
    "GIKKFLGSIWKFIKAFVKEIMN",
    # Short peptides
    "RRWQWR",
    "RWRWRW",
    "FKRIVQRIKDFL",
    "KFLKKAKKFGK",
    "GIGKFLHSAK",
    "KWKLFKKI",
    "RLFDKIRQ",
    # Longer peptides
    "GLFDIIKKIAESFLPKV",
    "GIGKFLHSAKKFGKAFV",
    "KWKLFKKIPKFLHLAK",
]
print(f"{len(MY_PEPTIDES)} peptides, length range: {min(len(s) for s in MY_PEPTIDES)}–{max(len(s) for s in MY_PEPTIDES)} aa")

25 peptides, length range: 6–37 aa

Build and run the pipeline¶

The dependency graph:

validate
   ├── predict_tm   ─┐
   └── predict_sol  ─┴── filter_tm >= 40°C ── rank top 15 by solubility

Both prediction stages run in parallel because they share the same dependency.

pipeline = DataPipeline(sequences=MY_PEPTIDES, verbose=True)

pipeline.add_filter(ValidAminoAcidFilter(), stage_name="validate")

pipeline.add_prediction(
    "temberture-regression", extractions="prediction",
    columns="melting_temperature", stage_name="predict_tm",
    depends_on=["validate"],
)
pipeline.add_prediction(
    "biolmsol", extractions="solubility_score",
    columns="solubility", stage_name="predict_sol",
    depends_on=["validate"],
)

pipeline.add_filter(ThresholdFilter("melting_temperature", min_value=40.0), stage_name="filter_tm")
pipeline.add_filter(RankingFilter("solubility", n=15, ascending=False), stage_name="top15")

pipeline.run()

Added stage: FilterStage('validate')
Added stage: PredictionStage('predict_tm', depends_on=['validate'])
Added stage: PredictionStage('predict_sol', depends_on=['validate'])
Added stage: FilterStage('filter_tm', depends_on=['predict_sol'])
Added stage: FilterStage('top15', depends_on=['filter_tm'])

############################################################
# Pipeline: DataPipeline
# Run ID: 20260617_141507_e9d0a416
# Initial sequences: 25
# Streaming: ENABLED
############################################################

Execution plan: 4 level(s)
  Level 1: validate
  Level 2: predict_tm, predict_sol (parallel)
  Level 3: filter_tm
  Level 4: top15

============================================================
Stage: validate
Input: 25 sequences
  Applying filter: ValidAminoAcidFilter(alphabet='ACDEFGHIKLMNPQRSTVWY')
  Filtered out: 0/25
  Remaining: 25

StageResult(validate: in=25, out=25, cached=0, computed=0, filtered=0, time=0.0s)
============================================================

Executing 2 stages in parallel...

============================================================
Stage: predict_tm
Input: 25 sequences
Depends on: validate
  Cached: 0/25
  To compute: 25
  Calling temberture-regression.predict...

============================================================
Stage: predict_sol
Input: 25 sequences
Depends on: validate
  Cached: 0/25
  To compute: 25
  Calling biolmsol.predict...

Completed: 25 sequences in 1 batches (max 5 concurrent)

StageResult(predict_sol: in=25, out=25, cached=0, computed=25, filtered=0, time=21.5s)
============================================================

Completed: 25 sequences in 1 batches (max 5 concurrent)

StageResult(predict_tm: in=25, out=25, cached=0, computed=25, filtered=0, time=26.0s)
============================================================

============================================================
Stage: filter_tm
Input: 25 sequences
Depends on: predict_sol
  Applying filter: ThresholdFilter(column='melting_temperature', min=40.0)
  Filtered out: 0/25
  Remaining: 25

StageResult(filter_tm: in=25, out=25, cached=0, computed=0, filtered=0, time=0.0s)
============================================================

============================================================
Stage: top15
Input: 25 sequences
Depends on: filter_tm
  Applying filter: RankingFilter(column='solubility', n=15, method='top')
  Filtered out: 10/25
  Remaining: 15

StageResult(top15: in=25, out=15, cached=0, computed=0, filtered=10, time=0.0s)
============================================================

############################################################
# Pipeline completed in 26.2s
# Final sequences: 15
############################################################

{'validate': StageResult(validate: in=25, out=25, cached=0, computed=0, filtered=0, time=0.0s),
 'predict_tm': StageResult(predict_tm: in=25, out=25, cached=0, computed=25, filtered=0, time=26.0s),
 'predict_sol': StageResult(predict_sol: in=25, out=25, cached=0, computed=25, filtered=0, time=21.5s),
 'filter_tm': StageResult(filter_tm: in=25, out=25, cached=0, computed=0, filtered=0, time=0.0s),
 'top15': StageResult(top15: in=25, out=15, cached=0, computed=0, filtered=10, time=0.0s)}

Explore results¶

pipeline.summary()

pipeline.stats()

# Top 10 sequences by melting temperature
pipeline.query("""
    SELECT s.sequence,
           MAX(CASE WHEN p.prediction_type = 'melting_temperature' THEN p.value END) AS tm,
           MAX(CASE WHEN p.prediction_type = 'solubility' THEN p.value END) AS solubility
    FROM sequences s
    JOIN predictions p ON s.sequence_id = p.sequence_id
    GROUP BY s.sequence
    ORDER BY tm DESC
    LIMIT 10
""")

pipeline.plot("funnel")

Next Steps¶

Check out additional tutorials at jupyter.biolm.ai, or head over to our BioLM Documentation to explore additional models and functionality.

	sequence	tm	solubility
0	LKLLKKLLKLLKKL	81.053093	1.993930
1	KLAKLAKKLAKLAK	79.101234	2.326550
2	GIKKFLGSIWKFIKAFVKEIMN	78.899384	0.614828
3	FKRIVQRIKDFL	74.703255	1.318316
4	KWKLFKKIPKFLHLAKKF	74.061577	1.513157
5	LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES	73.924812	1.391619
6	KWKLFKKI	73.292221	1.766447
7	KFLKKAKKFGK	70.871254	2.850865
8	KWKLFKKIPKFLHLAK	67.577049	1.308439
9	GLFDIIKKIAESFLPKV	67.185707	0.364214

All Examples

Launch a Full Peptide Screening Campaign, Model Properties in Parallel¶

Setup¶

Peptide library¶

Build and run the pipeline¶

Explore results¶

Next Steps¶

See more use-cases and APIs on your BioLM Console Catalog.¶

BioLM hosts deep learning models and runs inference at scale. You do the science.¶

Contact us to learn more.¶

Accelerate yourLead generation

	Stage	Input	Output	Filtered	Computed	Time (s)
0	validate	25	25	0	0	0.0
1	predict_tm	25	25	0	25	26.0
2	predict_sol	25	25	0	25	21.5
3	filter_tm	25	25	0	0	0.0
4	top15	25	15	10	0	0.0

	stage_name	status	input_count	output_count	completed_at
0	validate	completed	25	25	2026-06-17 14:15:07.468943
1	predict_sol	completed	25	25	2026-06-17 14:15:29.181349
2	predict_tm	completed	25	25	2026-06-17 14:15:33.469389
3	filter_tm	completed	25	25	2026-06-17 14:15:33.492115
4	top15	completed	25	15	2026-06-17 14:15:33.511078