Why Science? | bigLolly Genomic Lollipop Tracks

Science learning becomes powerful when students can tell a measured pattern from a model, a statistical decision from a mechanism, and a useful lead from a finished conclusion. UCSC bigLolly is a display mode for indexed bigBed records. Any suitable bigBed can be rendered as lollipops, with the ordinary BED score used for height by default or another numeric field selected through lollyField. A second field can control marker size through lollySizeField, lollyMaxSize can cap circles, and lollyNoStems can hide stems. The official example uses BED9 plus a negative-log p-value. A lollipop is a visual encoding rather than an evidence grade: its height, circle area and colour are meaningful only through the declared schema, transformation, scale and methods. This guide supports Primary Science, PSLE Science, Secondary Science, O-Level Science and STEM education as a longform evidence-reading route; it is not medical advice, a laboratory protocol, an admission promise or a career guarantee.

Related eduKate reading: bigBed indexed genome annotation tracks; BED and genomic feature coordinates; bigWig indexed genomic signal tracks; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.

Primary and official sources checked on 12 October 2026: UCSC Genome Browser bigLolly Track Format, checked 12 October 2026; UCSC Genome Browser bigBed Track Format, checked 12 October 2026; MOE G2 Computing Syllabus, checked 12 October 2026; 2026 Singapore–Cambridge O-Level Biology syllabus; MOE G2/G3 Lower Secondary Science teaching and learning syllabus; 2026 MOE G2 Computing syllabus. Specifications, tools and recommendations can change, so preserve dated documentation, exact source data, coordinate rules and software versions.

A deeper reading route

A lollipop plot compresses several visual signals into a small genomic space. Horizontal position anchors the record to the genome, vertical height can show one quantity, circle size can show another, and colour can express a category or display choice. That compactness is powerful, but it creates responsibility: readers must know which field drives each visual channel and whether the scale is linear, logarithmic or capped.

UCSC explains that bigLolly uses the indexed binary bigBed format and fetches only needed portions, caching them as a sparse file. It is therefore a display interpretation of a bigBed rather than a separate binary container. The underlying field schema remains authoritative; a browser setting selects how those stored values are rendered.

By default, lollipop height uses the BED score field. The lollyField setting can select another numeric extra field by name, while the official example schema adds pValueLog after BED9. This name-based selection depends on the autoSql declaration. Renaming or reordering columns without updating the schema can produce a plausible plot from the wrong measurement.

Marker size is independent of height when lollySizeField is set. That separation can show frequency and significance together, for example, but it can also invite perceptual error. Circle area grows with the square of radius, so an undocumented mapping can exaggerate differences. State the field, transformation, range and maximum-size cap in the visible legend.

lollyMaxSize limits the maximum lollipop size, and lollyNoStems can remove stems. These are display choices, not changes to source evidence. A capped circle no longer grows in proportion above the threshold, and a stemless point can obscure the exact baseline relationship. Figures should disclose caps and retain accessible numerical details.

The y-axis can be labelled and configured so readers know what height means. Negative values, missing values and very large values need deliberate rules. If the measurement is a negative base-ten logarithm, a higher stem means a smaller underlying p-value, not a larger probability of truth. A transformation must be named, not guessed from the shape.

Building follows the ordinary bigBed workflow: prepare sorted BED-compatible rows, supply an exact chromosome-size manifest, define extra fields in autoSql and run bedToBigBed with the matching type, such as the documented bed9+1 example. The resulting file can be interrogated with bigBedToBed, bigBedSummary and bigBedInfo before display settings are added.

UCSC notes that out-of-range coordinates cause conversion problems and mentions bedClip as a way to remove such rows. Clipping or dropping records may be technically convenient, but it changes scientific completeness. A responsible workflow quarantines invalid rows, reports their number and cause, and fixes the assembly or coordinate error rather than silently erasing evidence.

Overlapping records present another visual challenge. Several lollipops at the same or nearby position can hide one another, blend colours or imply one event. Aggregate only with a declared rule, and provide a details table or extract so every source record remains inspectable. Jitter or size changes should never create false genomic displacement.

For students, lollipops are a friendly introduction to multivariable graphs. The stick can show one number, the sweet can show another, and colour can show a category—all at a labelled genomic position. The important scientific habit is to decode the legend before telling a story about the tallest or largest symbol.

Biological claims remain separate from display. A high or large marker can summarise frequency, score, evidence or any other numeric field chosen by the producer. It does not automatically mean pathogenic, functional, causal or important. Interpretation needs the experiment, sample design, controls and an appropriate validation route.

A reproducible package includes source table, schema, assembly manifest, field dictionary, transformations, missing-value rules, conversion command, bigBed hash, track stanza, axis label, size cap and screenshot window. Tests should cover zero, negative, missing, tied, extreme and overlapping values so the plot remains honest at its edges.

Perceptual design should be tested with actual readers. People compare vertical position more accurately than circle area, and dense markers can appear more important simply because they occupy more ink. If height expresses the principal measurement, keep size secondary and bounded. If size is crucial, provide a table or tooltip with exact values so interpretation does not depend on estimating area by eye.

The genomic anchor also needs a declared convention. A one-base event may be represented as a one-base interval, while a structural or regional event can span many bases. The lollipop is usually drawn at a representative location, but that point should not erase the full interval. Keep start, end and any chosen centre rule available in the details view and methods note.

Statistical thresholds can produce discontinuous pictures. If only records below a p-value threshold are exported, a marker just outside the threshold disappears although its evidence may be nearly identical. Report selection rules and retain the complete tested universe or a suitable audit table. A lollipop plot is often the final layer of a filtering pipeline, not a neutral picture of all observations.

Version comparison should separate data from aesthetics. A new palette, axis maximum or size cap can make the same file look dramatically different; a new caller or sample set can change the records themselves. Store hashes for the source bigBed and track configuration independently. That makes visual regression tests meaningful and helps a later reader explain why a figure changed.

Export and citation deserve care as well. A static image should link back to the exact browser state or a durable methods record, because the picture alone omits tooltips, hidden records and configuration. Include a machine-readable table for accessibility and reuse. When the display supports a published claim, archive the source data and settings rather than relying on a live view that may later adopt new defaults.

A four-step claim ladder for this method

Stage	Question	Evidence	Responsible output
Place	Where does the item belong?	BED chromosome and coordinates	An assembly-specific genomic position
Raise	Which field controls vertical height?	score or the named lollyField	A documented y-axis value
Size	Which field controls marker size?	lollySizeField and lollyMaxSize	A bounded second encoding
Judge	What does the symbol support?	Schema, transformation, samples and validation	A qualified visual claim